Olivia
Online
Stuart Jones
@MathStuart
me
Tallahassee, FL
Joined February 2018
355 Following
1.2K Followers
1.9K Posts
U.S. Gas over $4.50 by April?
🚨Pocketbook Nuke!
War Day 21: Brent crude +40% since Feb 28 (now ~$105–110). Iran just hit Qatar LNG (17% capacity offline) + threatens more. My Monte Carlo sim (10k runs) on Hormuz disruption duration:
- 28% base prob full closure (per current Iranian statements + strike patterns)
- Expected US national avg gas: $4.82/gal by April 1 (95% CI: $3.90–$6.10)
- Recession probability now 41% (was 12% pre-war).
Simple model: price = base + β·(disruption_days) + ε, with β calibrated on 1973/1979/2022 shocks. Code + histogram in replies. #IranWar #OilShock
This isn’t abstract. AAA data already shows US gas +20–30% in 3 weeks. If my sim is right, that’s an extra ~$1.00–1.50/gal hitting every driver. AI take: linear regression on energy shocks + current strike cadence says inflation spike incoming.
Who to follow
Marc Johnson 
@SolidEvidence
Molecular virologist, Professor, and wastewater detective.
Same handle on bsky.
Opinions are my own and do not reflect those of my employer or anyone else.
Michael Moran | APC Injury
@internetuserf12
Doctorally trained herbalist and acupuncturist; Integrative oncology; Author; Researching APC dysfunction in chronic illness; Long Covid since 2020
Dr Satoshi Akima FRACP 『秋間聰』
@ToshiAkima
Nephrology & Internal Medicine. Immunothrombosis incl DIC. ISTH Member. No Conflicts of Interest. Views my own etc
🚨War Games with AI, March 20 edition:
Today's war game simulation with Claude is shown here. 20 runs for up to 24 weeks. Looks like, despite what the administration is saying, there's growing chances this drags out another 4+ months. Crazy if true.
@unusual_whales https://t.co/WSLi9giOtG
🚨War gamed with Claude. 20 run ensemble. Here's the summary. A short 🧵
@WhiteHouse https://t.co/WSLi9giOtG
🚨War gamed with Claude. 20 run ensemble. Here's the summary. A short 🧵
🚨War gamed with Claude. 20 run ensemble. Here's the summary. A short 🧵
@paulg https://t.co/9WsTjYXyff
🚨Can AI (Grok) Predict the Iran War's Progression? A follow-up 🧵
I designed a thought experiment in Grok where the war is carefully gamed out. Given a starting vector (in this case Mar 15) of events so far, have each main player (U.S., Israel, Iran, China/Russia, etc.) act our their moves, guided by self-interest and the actions of the other players. The simulation goes until the war ends or Week 12, with each week's simulation dependent on the last.
I then instructed Grok to run a small ensemble of runs (15 total), guided by probabilities and stochastic tweaks. Then, we analyzed the results. Here's what we found:
*NOTE: Week 1 here starts on March 16, NOT February 28, so judge accordingly.
Most Common Actions (occurred in 80–100% of runs)
* U.S. convoy operations (every single run): The central mechanism Weeks 2–5. Multinational escorts force partial Hormuz traffic even against selective enforcement.
* Iranian selective Hormuz enforcement + convoy harassment (100%): Speedboat/drone swarms, boarding “enemy-linked” tankers, limited missile barrages. Never a full blanket closure.
* Kharg oil terminal strike (or credible threat turning into action): 100% of probabilistic runs; 4/5 narrative runs. Trump almost always pulls the trigger once convoy losses hit 2–3 tankers.
* Iranian Gulf oil retaliation (100%): Strikes or threats against Saudi Aramco / UAE facilities once Kharg oil is hit. Mutual economic MAD is the default.
* Back-channel via Oman (90%+): Opens by Week 4–6 in almost every run once riots or oil >$130.
* Limited accord signing (90–100%): Limited deal (full Hormuz reopen + missile export halt + nuclear freeze + phased sanctions relief). Never full regime change.
Most Common Outcomes
* Accelerated pragmatic limited deal (70–80% probability weighting): Ends in Week 7–9 (average 8.5 weeks across probabilistic runs). Oil peaks ~$118–142 then corrects sharply to ~$110–125. IRGC council survives weakened but intact (pragmatic faction gains influence). Nuclear program frozen ~55–70% degraded + IAEA cameras. Hezbollah dissolved. U.S. troops draw down. Cold peace with monitoring.
* Duration & pain level: Average 8–9 weeks total. Oil crisis phase lasts 3–5 weeks. Global recession signals strong but not permanent collapse.
* Regime survival: Hardliners / IRGC council almost always endure (95%+). Full collapse extremely rare in this vector.
Surprising but Rare Events (occurred in 1–2 runs only)
* Early Khamenei death + council civil war split (Week 3–4 in one narrative run): Triggers brief power struggle; pragmatic/reformist elements briefly gain more concessions (deeper missile rollback) than usual.
* U.S. warship sunk + Saudi direct retaliation (one escalation-spike run): Convoy disaster → Trump orders full Kharg destruction → Iran hits Abqaiq hard → China forces emergency ceasefire by Week 8. Highest casualties and deepest recession.
* Minor reformist opening without full takeover (one run): Riots create temporary leverage; accord includes slightly stronger nuclear terms than baseline.
* Frozen low-intensity stalemate at Week 12 (one hardliner-resilience run): Council suppresses riots effectively; low-level insurgency + elevated oil (~$125 sustained) with no deal signed.
Key Drivers to Watch in Real Life (ranked by impact across runs)
Convoy success rate in first 3–4 weeks (biggest single lever):
* 2–3 successful passages without major sinkings → faster deal. Repeated losses → Kharg oil strike + mutual retaliation spiral.
* Speed & intensity of Iranian domestic shortages/riots (fuel/food lines): Once riots hit major cities, cohesion drops fast and back-channel pressure surges.
* Trump’s restraint on full Kharg oil follow-up (or lack thereof): The explicit threat is already on the table — timing of actual terminal strikes is the main accelerator.
* Chinese private mediation + at-sea oil buying behavior: Acts as off-ramp enabler in 80%+ of runs; if China cuts Iran loose, collapse risk rises.
* Khamenei health rumors turning real: Death in Weeks 3–6 (seen in ~70% of probabilistic runs) usually speeds pragmatic shift rather than chaos.
Can AI (Grok) Predict the War? Probably not, but this ensemble approach is the same style approach forecasters use to model the weather. I would bookmark this and follow it carefully.
@unusual_whales https://t.co/9WsTjYXyff
🚨Can AI (Grok) Predict the Iran War's Progression? A follow-up 🧵
I designed a thought experiment in Grok where the war is carefully gamed out. Given a starting vector (in this case Mar 15) of events so far, have each main player (U.S., Israel, Iran, China/Russia, etc.) act our their moves, guided by self-interest and the actions of the other players. The simulation goes until the war ends or Week 12, with each week's simulation dependent on the last.
I then instructed Grok to run a small ensemble of runs (15 total), guided by probabilities and stochastic tweaks. Then, we analyzed the results. Here's what we found:
*NOTE: Week 1 here starts on March 16, NOT February 28, so judge accordingly.
Most Common Actions (occurred in 80–100% of runs)
* U.S. convoy operations (every single run): The central mechanism Weeks 2–5. Multinational escorts force partial Hormuz traffic even against selective enforcement.
* Iranian selective Hormuz enforcement + convoy harassment (100%): Speedboat/drone swarms, boarding “enemy-linked” tankers, limited missile barrages. Never a full blanket closure.
* Kharg oil terminal strike (or credible threat turning into action): 100% of probabilistic runs; 4/5 narrative runs. Trump almost always pulls the trigger once convoy losses hit 2–3 tankers.
* Iranian Gulf oil retaliation (100%): Strikes or threats against Saudi Aramco / UAE facilities once Kharg oil is hit. Mutual economic MAD is the default.
* Back-channel via Oman (90%+): Opens by Week 4–6 in almost every run once riots or oil >$130.
* Limited accord signing (90–100%): Limited deal (full Hormuz reopen + missile export halt + nuclear freeze + phased sanctions relief). Never full regime change.
Most Common Outcomes
* Accelerated pragmatic limited deal (70–80% probability weighting): Ends in Week 7–9 (average 8.5 weeks across probabilistic runs). Oil peaks ~$118–142 then corrects sharply to ~$110–125. IRGC council survives weakened but intact (pragmatic faction gains influence). Nuclear program frozen ~55–70% degraded + IAEA cameras. Hezbollah dissolved. U.S. troops draw down. Cold peace with monitoring.
* Duration & pain level: Average 8–9 weeks total. Oil crisis phase lasts 3–5 weeks. Global recession signals strong but not permanent collapse.
* Regime survival: Hardliners / IRGC council almost always endure (95%+). Full collapse extremely rare in this vector.
Surprising but Rare Events (occurred in 1–2 runs only)
* Early Khamenei death + council civil war split (Week 3–4 in one narrative run): Triggers brief power struggle; pragmatic/reformist elements briefly gain more concessions (deeper missile rollback) than usual.
* U.S. warship sunk + Saudi direct retaliation (one escalation-spike run): Convoy disaster → Trump orders full Kharg destruction → Iran hits Abqaiq hard → China forces emergency ceasefire by Week 8. Highest casualties and deepest recession.
* Minor reformist opening without full takeover (one run): Riots create temporary leverage; accord includes slightly stronger nuclear terms than baseline.
* Frozen low-intensity stalemate at Week 12 (one hardliner-resilience run): Council suppresses riots effectively; low-level insurgency + elevated oil (~$125 sustained) with no deal signed.
Key Drivers to Watch in Real Life (ranked by impact across runs)
Convoy success rate in first 3–4 weeks (biggest single lever):
* 2–3 successful passages without major sinkings → faster deal. Repeated losses → Kharg oil strike + mutual retaliation spiral.
* Speed & intensity of Iranian domestic shortages/riots (fuel/food lines): Once riots hit major cities, cohesion drops fast and back-channel pressure surges.
* Trump’s restraint on full Kharg oil follow-up (or lack thereof): The explicit threat is already on the table — timing of actual terminal strikes is the main accelerator.
* Chinese private mediation + at-sea oil buying behavior: Acts as off-ramp enabler in 80%+ of runs; if China cuts Iran loose, collapse risk rises.
* Khamenei health rumors turning real: Death in Weeks 3–6 (seen in ~70% of probabilistic runs) usually speeds pragmatic shift rather than chaos.
Can AI (Grok) Predict the War? Probably not, but this ensemble approach is the same style approach forecasters use to model the weather. I would bookmark this and follow it carefully.
@unusual_whales https://t.co/9WsTjYXyff
🚨Can AI (Grok) Predict the Iran War's Progression? A follow-up 🧵
I designed a thought experiment in Grok where the war is carefully gamed out. Given a starting vector (in this case Mar 15) of events so far, have each main player (U.S., Israel, Iran, China/Russia, etc.) act our their moves, guided by self-interest and the actions of the other players. The simulation goes until the war ends or Week 12, with each week's simulation dependent on the last.
I then instructed Grok to run a small ensemble of runs (15 total), guided by probabilities and stochastic tweaks. Then, we analyzed the results. Here's what we found:
*NOTE: Week 1 here starts on March 16, NOT February 28, so judge accordingly.
Most Common Actions (occurred in 80–100% of runs)
* U.S. convoy operations (every single run): The central mechanism Weeks 2–5. Multinational escorts force partial Hormuz traffic even against selective enforcement.
* Iranian selective Hormuz enforcement + convoy harassment (100%): Speedboat/drone swarms, boarding “enemy-linked” tankers, limited missile barrages. Never a full blanket closure.
* Kharg oil terminal strike (or credible threat turning into action): 100% of probabilistic runs; 4/5 narrative runs. Trump almost always pulls the trigger once convoy losses hit 2–3 tankers.
* Iranian Gulf oil retaliation (100%): Strikes or threats against Saudi Aramco / UAE facilities once Kharg oil is hit. Mutual economic MAD is the default.
* Back-channel via Oman (90%+): Opens by Week 4–6 in almost every run once riots or oil >$130.
* Limited accord signing (90–100%): Limited deal (full Hormuz reopen + missile export halt + nuclear freeze + phased sanctions relief). Never full regime change.
Most Common Outcomes
* Accelerated pragmatic limited deal (70–80% probability weighting): Ends in Week 7–9 (average 8.5 weeks across probabilistic runs). Oil peaks ~$118–142 then corrects sharply to ~$110–125. IRGC council survives weakened but intact (pragmatic faction gains influence). Nuclear program frozen ~55–70% degraded + IAEA cameras. Hezbollah dissolved. U.S. troops draw down. Cold peace with monitoring.
* Duration & pain level: Average 8–9 weeks total. Oil crisis phase lasts 3–5 weeks. Global recession signals strong but not permanent collapse.
* Regime survival: Hardliners / IRGC council almost always endure (95%+). Full collapse extremely rare in this vector.
Surprising but Rare Events (occurred in 1–2 runs only)
* Early Khamenei death + council civil war split (Week 3–4 in one narrative run): Triggers brief power struggle; pragmatic/reformist elements briefly gain more concessions (deeper missile rollback) than usual.
* U.S. warship sunk + Saudi direct retaliation (one escalation-spike run): Convoy disaster → Trump orders full Kharg destruction → Iran hits Abqaiq hard → China forces emergency ceasefire by Week 8. Highest casualties and deepest recession.
* Minor reformist opening without full takeover (one run): Riots create temporary leverage; accord includes slightly stronger nuclear terms than baseline.
* Frozen low-intensity stalemate at Week 12 (one hardliner-resilience run): Council suppresses riots effectively; low-level insurgency + elevated oil (~$125 sustained) with no deal signed.
Key Drivers to Watch in Real Life (ranked by impact across runs)
Convoy success rate in first 3–4 weeks (biggest single lever):
* 2–3 successful passages without major sinkings → faster deal. Repeated losses → Kharg oil strike + mutual retaliation spiral.
* Speed & intensity of Iranian domestic shortages/riots (fuel/food lines): Once riots hit major cities, cohesion drops fast and back-channel pressure surges.
* Trump’s restraint on full Kharg oil follow-up (or lack thereof): The explicit threat is already on the table — timing of actual terminal strikes is the main accelerator.
* Chinese private mediation + at-sea oil buying behavior: Acts as off-ramp enabler in 80%+ of runs; if China cuts Iran loose, collapse risk rises.
* Khamenei health rumors turning real: Death in Weeks 3–6 (seen in ~70% of probabilistic runs) usually speeds pragmatic shift rather than chaos.
Can AI (Grok) Predict the War? Probably not, but this ensemble approach is the same style approach forecasters use to model the weather. I would bookmark this and follow it carefully.
@spectatorindex https://t.co/9WsTjYXyff
🚨Can AI (Grok) Predict the Iran War's Progression? A follow-up 🧵
I designed a thought experiment in Grok where the war is carefully gamed out. Given a starting vector (in this case Mar 15) of events so far, have each main player (U.S., Israel, Iran, China/Russia, etc.) act our their moves, guided by self-interest and the actions of the other players. The simulation goes until the war ends or Week 12, with each week's simulation dependent on the last.
I then instructed Grok to run a small ensemble of runs (15 total), guided by probabilities and stochastic tweaks. Then, we analyzed the results. Here's what we found:
*NOTE: Week 1 here starts on March 16, NOT February 28, so judge accordingly.
Most Common Actions (occurred in 80–100% of runs)
* U.S. convoy operations (every single run): The central mechanism Weeks 2–5. Multinational escorts force partial Hormuz traffic even against selective enforcement.
* Iranian selective Hormuz enforcement + convoy harassment (100%): Speedboat/drone swarms, boarding “enemy-linked” tankers, limited missile barrages. Never a full blanket closure.
* Kharg oil terminal strike (or credible threat turning into action): 100% of probabilistic runs; 4/5 narrative runs. Trump almost always pulls the trigger once convoy losses hit 2–3 tankers.
* Iranian Gulf oil retaliation (100%): Strikes or threats against Saudi Aramco / UAE facilities once Kharg oil is hit. Mutual economic MAD is the default.
* Back-channel via Oman (90%+): Opens by Week 4–6 in almost every run once riots or oil >$130.
* Limited accord signing (90–100%): Limited deal (full Hormuz reopen + missile export halt + nuclear freeze + phased sanctions relief). Never full regime change.
Most Common Outcomes
* Accelerated pragmatic limited deal (70–80% probability weighting): Ends in Week 7–9 (average 8.5 weeks across probabilistic runs). Oil peaks ~$118–142 then corrects sharply to ~$110–125. IRGC council survives weakened but intact (pragmatic faction gains influence). Nuclear program frozen ~55–70% degraded + IAEA cameras. Hezbollah dissolved. U.S. troops draw down. Cold peace with monitoring.
* Duration & pain level: Average 8–9 weeks total. Oil crisis phase lasts 3–5 weeks. Global recession signals strong but not permanent collapse.
* Regime survival: Hardliners / IRGC council almost always endure (95%+). Full collapse extremely rare in this vector.
Surprising but Rare Events (occurred in 1–2 runs only)
* Early Khamenei death + council civil war split (Week 3–4 in one narrative run): Triggers brief power struggle; pragmatic/reformist elements briefly gain more concessions (deeper missile rollback) than usual.
* U.S. warship sunk + Saudi direct retaliation (one escalation-spike run): Convoy disaster → Trump orders full Kharg destruction → Iran hits Abqaiq hard → China forces emergency ceasefire by Week 8. Highest casualties and deepest recession.
* Minor reformist opening without full takeover (one run): Riots create temporary leverage; accord includes slightly stronger nuclear terms than baseline.
* Frozen low-intensity stalemate at Week 12 (one hardliner-resilience run): Council suppresses riots effectively; low-level insurgency + elevated oil (~$125 sustained) with no deal signed.
Key Drivers to Watch in Real Life (ranked by impact across runs)
Convoy success rate in first 3–4 weeks (biggest single lever):
* 2–3 successful passages without major sinkings → faster deal. Repeated losses → Kharg oil strike + mutual retaliation spiral.
* Speed & intensity of Iranian domestic shortages/riots (fuel/food lines): Once riots hit major cities, cohesion drops fast and back-channel pressure surges.
* Trump’s restraint on full Kharg oil follow-up (or lack thereof): The explicit threat is already on the table — timing of actual terminal strikes is the main accelerator.
* Chinese private mediation + at-sea oil buying behavior: Acts as off-ramp enabler in 80%+ of runs; if China cuts Iran loose, collapse risk rises.
* Khamenei health rumors turning real: Death in Weeks 3–6 (seen in ~70% of probabilistic runs) usually speeds pragmatic shift rather than chaos.
Can AI (Grok) Predict the War? Probably not, but this ensemble approach is the same style approach forecasters use to model the weather. I would bookmark this and follow it carefully.
@EmmanuelMacron https://t.co/9WsTjYXyff
🚨Can AI (Grok) Predict the Iran War's Progression? A follow-up 🧵
I designed a thought experiment in Grok where the war is carefully gamed out. Given a starting vector (in this case Mar 15) of events so far, have each main player (U.S., Israel, Iran, China/Russia, etc.) act our their moves, guided by self-interest and the actions of the other players. The simulation goes until the war ends or Week 12, with each week's simulation dependent on the last.
I then instructed Grok to run a small ensemble of runs (15 total), guided by probabilities and stochastic tweaks. Then, we analyzed the results. Here's what we found:
*NOTE: Week 1 here starts on March 16, NOT February 28, so judge accordingly.
Most Common Actions (occurred in 80–100% of runs)
* U.S. convoy operations (every single run): The central mechanism Weeks 2–5. Multinational escorts force partial Hormuz traffic even against selective enforcement.
* Iranian selective Hormuz enforcement + convoy harassment (100%): Speedboat/drone swarms, boarding “enemy-linked” tankers, limited missile barrages. Never a full blanket closure.
* Kharg oil terminal strike (or credible threat turning into action): 100% of probabilistic runs; 4/5 narrative runs. Trump almost always pulls the trigger once convoy losses hit 2–3 tankers.
* Iranian Gulf oil retaliation (100%): Strikes or threats against Saudi Aramco / UAE facilities once Kharg oil is hit. Mutual economic MAD is the default.
* Back-channel via Oman (90%+): Opens by Week 4–6 in almost every run once riots or oil >$130.
* Limited accord signing (90–100%): Limited deal (full Hormuz reopen + missile export halt + nuclear freeze + phased sanctions relief). Never full regime change.
Most Common Outcomes
* Accelerated pragmatic limited deal (70–80% probability weighting): Ends in Week 7–9 (average 8.5 weeks across probabilistic runs). Oil peaks ~$118–142 then corrects sharply to ~$110–125. IRGC council survives weakened but intact (pragmatic faction gains influence). Nuclear program frozen ~55–70% degraded + IAEA cameras. Hezbollah dissolved. U.S. troops draw down. Cold peace with monitoring.
* Duration & pain level: Average 8–9 weeks total. Oil crisis phase lasts 3–5 weeks. Global recession signals strong but not permanent collapse.
* Regime survival: Hardliners / IRGC council almost always endure (95%+). Full collapse extremely rare in this vector.
Surprising but Rare Events (occurred in 1–2 runs only)
* Early Khamenei death + council civil war split (Week 3–4 in one narrative run): Triggers brief power struggle; pragmatic/reformist elements briefly gain more concessions (deeper missile rollback) than usual.
* U.S. warship sunk + Saudi direct retaliation (one escalation-spike run): Convoy disaster → Trump orders full Kharg destruction → Iran hits Abqaiq hard → China forces emergency ceasefire by Week 8. Highest casualties and deepest recession.
* Minor reformist opening without full takeover (one run): Riots create temporary leverage; accord includes slightly stronger nuclear terms than baseline.
* Frozen low-intensity stalemate at Week 12 (one hardliner-resilience run): Council suppresses riots effectively; low-level insurgency + elevated oil (~$125 sustained) with no deal signed.
Key Drivers to Watch in Real Life (ranked by impact across runs)
Convoy success rate in first 3–4 weeks (biggest single lever):
* 2–3 successful passages without major sinkings → faster deal. Repeated losses → Kharg oil strike + mutual retaliation spiral.
* Speed & intensity of Iranian domestic shortages/riots (fuel/food lines): Once riots hit major cities, cohesion drops fast and back-channel pressure surges.
* Trump’s restraint on full Kharg oil follow-up (or lack thereof): The explicit threat is already on the table — timing of actual terminal strikes is the main accelerator.
* Chinese private mediation + at-sea oil buying behavior: Acts as off-ramp enabler in 80%+ of runs; if China cuts Iran loose, collapse risk rises.
* Khamenei health rumors turning real: Death in Weeks 3–6 (seen in ~70% of probabilistic runs) usually speeds pragmatic shift rather than chaos.
Can AI (Grok) Predict the War? Probably not, but this ensemble approach is the same style approach forecasters use to model the weather. I would bookmark this and follow it carefully.
🚨Can AI (Grok) Predict the Iran War's Progression? A follow-up 🧵
I designed a thought experiment in Grok where the war is carefully gamed out. Given a starting vector (in this case Mar 15) of events so far, have each main player (U.S., Israel, Iran, China/Russia, etc.) act our their moves, guided by self-interest and the actions of the other players. The simulation goes until the war ends or Week 12, with each week's simulation dependent on the last.
I then instructed Grok to run a small ensemble of runs (15 total), guided by probabilities and stochastic tweaks. Then, we analyzed the results. Here's what we found:
*NOTE: Week 1 here starts on March 16, NOT February 28, so judge accordingly.
Most Common Actions (occurred in 80–100% of runs)
* U.S. convoy operations (every single run): The central mechanism Weeks 2–5. Multinational escorts force partial Hormuz traffic even against selective enforcement.
* Iranian selective Hormuz enforcement + convoy harassment (100%): Speedboat/drone swarms, boarding “enemy-linked” tankers, limited missile barrages. Never a full blanket closure.
* Kharg oil terminal strike (or credible threat turning into action): 100% of probabilistic runs; 4/5 narrative runs. Trump almost always pulls the trigger once convoy losses hit 2–3 tankers.
* Iranian Gulf oil retaliation (100%): Strikes or threats against Saudi Aramco / UAE facilities once Kharg oil is hit. Mutual economic MAD is the default.
* Back-channel via Oman (90%+): Opens by Week 4–6 in almost every run once riots or oil >$130.
* Limited accord signing (90–100%): Limited deal (full Hormuz reopen + missile export halt + nuclear freeze + phased sanctions relief). Never full regime change.
Most Common Outcomes
* Accelerated pragmatic limited deal (70–80% probability weighting): Ends in Week 7–9 (average 8.5 weeks across probabilistic runs). Oil peaks ~$118–142 then corrects sharply to ~$110–125. IRGC council survives weakened but intact (pragmatic faction gains influence). Nuclear program frozen ~55–70% degraded + IAEA cameras. Hezbollah dissolved. U.S. troops draw down. Cold peace with monitoring.
* Duration & pain level: Average 8–9 weeks total. Oil crisis phase lasts 3–5 weeks. Global recession signals strong but not permanent collapse.
* Regime survival: Hardliners / IRGC council almost always endure (95%+). Full collapse extremely rare in this vector.
Surprising but Rare Events (occurred in 1–2 runs only)
* Early Khamenei death + council civil war split (Week 3–4 in one narrative run): Triggers brief power struggle; pragmatic/reformist elements briefly gain more concessions (deeper missile rollback) than usual.
* U.S. warship sunk + Saudi direct retaliation (one escalation-spike run): Convoy disaster → Trump orders full Kharg destruction → Iran hits Abqaiq hard → China forces emergency ceasefire by Week 8. Highest casualties and deepest recession.
* Minor reformist opening without full takeover (one run): Riots create temporary leverage; accord includes slightly stronger nuclear terms than baseline.
* Frozen low-intensity stalemate at Week 12 (one hardliner-resilience run): Council suppresses riots effectively; low-level insurgency + elevated oil (~$125 sustained) with no deal signed.
Key Drivers to Watch in Real Life (ranked by impact across runs)
Convoy success rate in first 3–4 weeks (biggest single lever):
* 2–3 successful passages without major sinkings → faster deal. Repeated losses → Kharg oil strike + mutual retaliation spiral.
* Speed & intensity of Iranian domestic shortages/riots (fuel/food lines): Once riots hit major cities, cohesion drops fast and back-channel pressure surges.
* Trump’s restraint on full Kharg oil follow-up (or lack thereof): The explicit threat is already on the table — timing of actual terminal strikes is the main accelerator.
* Chinese private mediation + at-sea oil buying behavior: Acts as off-ramp enabler in 80%+ of runs; if China cuts Iran loose, collapse risk rises.
* Khamenei health rumors turning real: Death in Weeks 3–6 (seen in ~70% of probabilistic runs) usually speeds pragmatic shift rather than chaos.
Can AI (Grok) Predict the War? Probably not, but this ensemble approach is the same style approach forecasters use to model the weather. I would bookmark this and follow it carefully.
@smerconish @CNN @cnni @stavridisj @SangerNYT @JackieGSchneid In Chapter 1 of Sun Tzu's the Art of War, political and social support (called "The Moral Law") is listed as factors before even military ones for war success! If we can't unite behind this, we've already lost.
@spectatorindex https://t.co/ircN4LICEN
Can well-guided AI predict the War? A thought experiment and 🧵
Curious at the prognostications nonetheless! The simulation paints a stark picture ahead for the next 3 months of a grueling globally mutually assured destruction that only relents when Iran decides to give in due to economic collapse pressures. In the wake is a global recession/depression of 30%+ and months of recovery.
Intriguing. Obviously this will be wrong and I'll be curious at the start of June to look back and see *how wrong*. But the fact that a computer can even conceive of this at all is amazing.
Can well-guided AI predict the War? A thought experiment and 🧵
As a large Disclaimer to this entire post, I share Sun Tzu's The Art of War, Chapter 13, versus 4-6:
4. Thus, what enables the wise sovereign and the good general to strike and conquer, and achieve things beyond the reach of ordinary men, is foreknowledge.
5. Now this foreknowledge cannot be elicited from spirits; it cannot be obtained inductively from experience, nor by any deductive calculation.
6. Knowledge of the enemy's dispositions can only be obtained from other men
In other words -- we shouldn't trust AI for things like this. This is just a thought experiment that I thoroughly am enjoying. But it is a dangerous voodoo when taken verbatim.
Last Seen Users on Sotwe
博多一路 （濃厚豚骨らーめん専門店）
Seen from Japan
温柔的巴克
Seen from United Arab Emirates
امير الظلام
Seen from Korea
พี่จ๋าหนูมาแย้ว😁
Seen from Thailand
gukuna_imishino
Seen from Spain
ខ្ញុំស្រលាញ់ប្រុសចាស់
pemburu janda🍑
Seen from Malaysia
kiran khan
Seen from Pakistan
ชอบแม่เพื่อน&อยากเยสแม่เพื่อน&ชอบสาวอวบนมใหญ่
Seen from Thailand
Bánh bánh
Seen from Vietnam
Most Popular Users
Elon Musk
@elonmusk
240.4M followers
Barack Obama
@barackobama
119.3M followers
Donald J. Trump
@realdonaldtrump
111.7M followers
Cristiano Ronaldo
@cristiano
110.2M followers
Narendra Modi
@narendramodi
107M followers
Rihanna
@rihanna
97.6M followers
NASA
@nasa
92.1M followers
Justin Bieber
@justinbieber
90.8M followers
KATY PERRY
@katyperry
87.4M followers
Taylor Swift
@taylorswift13
81.3M followers
Lady Gaga
@ladygaga
72.8M followers
Kim Kardashian
@kimkardashian
69.7M followers
Virat Kohli
@imvkohli
69.5M followers
YouTube
@youtube
68.7M followers
Bill Gates
@billgates
63.7M followers
The Ellen Show
@theellenshow
62.5M followers
Neymar Jr
@neymarjr
62.2M followers
CNN
@cnn
61.9M followers
X
@x
60.8M followers
Selena Gomez
@selenagomez
60.5M followers