Geopolitical Conflict
Ukrainian drones attacked Russian industrial sites across multiple regions Wednesday, sparking a fire at Astrakhan's gas processing plant despite Russia's claim of intercepting 286 drones, while Russia launched 139 drones at Ukraine overnight.(The Moscow Times)Russia · Ukraine
The Brief · Wednesday, May 13, 2026
Palestine – Daily Risk Brief
Palestine remains in critical instability (score: 34.2) with sustained Israeli military operations in Gaza and West Bank settler violence driving active conflict escalation. High confidence: kinetic operations ongoing as of 12 May 2026, with reconstruction costs estimated at $71 billion and no political settlement framework evident.
May 12, 2026 · Score 34.2
39 citationsCountry · DailyPalestineICD 203?ICD 203 — Analytic TradecraftThe US Intelligence Community’s Intelligence Community Directive 203defines the standards every serious analyst brief follows: sourced claims, calibrated probability language (“likely” / “highly likely” rather than guessing), explicit confidence levels, and alternative-analysis sections that name what could prove the brief wrong.
GeoMemo briefs are generated under an ICD 203 prompt: bottom line up-front, every assertion cited to the dispatch it came from, and probability framing that survives analyst scrutiny.
GeoMemo briefs are generated under an ICD 203 prompt: bottom line up-front, every assertion cited to the dispatch it came from, and probability framing that survives analyst scrutiny.
Newsletter · free
Start your morning with the GeoMemo Brief.
Never miss what’s moving. Ten minutes. Multi-capital lens. In your inbox by 07:30 EDT.
Free · no upsells · unsubscribe in one click
BLUF signal · Palestine · events 90d
1399events · 90 daily points
2026-02-13→2026-05-13
Latest dispatchesAll dispatches →
Defense & Arms Transfers
EU Defence Commissioner Andrius Kubilius urged Eastern European nations to shape Europe's emerging independent defence architecture amid a historic strategic shift away from US-led NATO toward autonomous European collective defence capability.(Euractiv)Lithuania · Spain · GermanyGeopolitical Conflict
Analyst Robert Kagan argues in The Atlantic that the U.S. and Israel face strategic defeat against Iran, having lost effective control of the Strait of Hormuz despite military efforts, fundamentally undermining American global hegemony and energy security.(Ettela'at Online)United States · Iran · IsraelInternational Relations
South Korea's security chief said the nation is reviewing potential military roles in U.S.-led Hormuz initiatives following an unexplained explosion aboard a Korean cargo ship, with possible contributions ranging from minimal to substantial involvement.(Korea JoongAng Daily)South Korea · United States · IranGeopolitical Economics
China's liquefied natural gas imports rebounded from an eight-year low in April to their highest level since late February, though remaining below five-year averages, as Middle Eastern supply disruptions and rising summer cooling demand drove global prices higher.(Investing.com India)China · Qatar · IranGeopolitical Economics
United Airlines will resume daily direct flights between Houston and Caracas starting August 11 after nearly nine years, using Boeing 737 MAX 8 aircraft, benefiting Venezuelan Texans and supporting US-Venezuela energy sector cooperation.(La Nacion — LA NACION)United States · VenezuelaInternational Relations
Nigerian billionaire Tony Elumelu urged Africa to abandon victimhood and welcome foreign investment from all nations, arguing that infrastructure development and job creation for Africa's young population should take priority over historical grievances.(The Hindu)Nigeria · France · KenyaInternational Relations
China altered the official transliteration of Secretary of State Marco Rubio's surname in Mandarin to circumvent existing sanctions against him, allowing his entry despite prior restrictions, amid escalating US-China tensions over Iran and the Strait of Hormuz.(La Nacion — LA NACION)United States · China · IranMore:NDTV
Geopolitical Economics
Trump traveled to Beijing with over 15 tech executives including Elon Musk and Tim Cook, seeking commercial access from Xi Jinping. Last-minute invitee Jensen Huang joined the delegation after an Alaska stopover, as Nvidia seeks regulatory approval to sell advanced AI chips to China.(La Nacion — LA NACION)United States · ChinaGeopolitical Economics
Argentina's sunflower production is projected to reach eight million tons in 2026-2027, a 14.2 percent increase, while first-quarter exports generated over one billion dollars, nearly doubling year-over-year revenues amid rising global demand.(La Nacion — LA NACION)Argentina · United States · UkraineThree judgments you need to make today?How these are pickedFrom the full board of Bayesian-tracked threats, we rank by |7-day probability change| × max(current probability, 0.30). That weighting keeps a high-probability threat that moved 3pp above a noisy 10%-probability threat that moved 8pp. Top three picks = what moved most, among the threats that actually matter today.weighted movement · 14 days probability trajectory
01 · nuclear · North Korea
North Korea provocations
The revelation of a Russian cargo ship carrying nuclear reactor components for North Korea suggests a potential escalation of nuclear capabilities, which could increase the likelihood of North Korean provocations.
Russian cargo ship sunk off Spain carried nuclear reactors for North Korea, probe reveals
-0.0σVS 82D MEAN 71%?What this σ number meansIn plain English: is today’s probability unusual compared to recent history, or business-as-usual?
±1σ (typical) — inside the normal range. Baseline noise.
±1σ to ±2σ (elevated) — outside the usual range. Worth a second look.
Beyond ±2σ (critical) — statistically unusual. Something changed.
Technical: σ (lowercase sigma, the standard-deviation symbol — not to be confused with Σ, the summation symbol) = (current probability − 60-day baseline mean) / baseline standard deviation. Baseline excludes the last 7 days so recent movement isn’t self-referenced.
±1σ (typical) — inside the normal range. Baseline noise.
±1σ to ±2σ (elevated) — outside the usual range. Worth a second look.
Beyond ±2σ (critical) — statistically unusual. Something changed.
Technical: σ (lowercase sigma, the standard-deviation symbol — not to be confused with Σ, the summation symbol) = (current probability − 60-day baseline mean) / baseline standard deviation. Baseline excludes the last 7 days so recent movement isn’t self-referenced.
Geographic scope?Geographic scopeWhere the dispatches that move this threat’s probability are set. We extract the country codes each article references (from NER + its source publication’s country field), aggregate them across all {evidence} pieces that have touched this threat, and show the top 5 by share.
Why it matters: a threat where the coverage concentrates in one country reads differently from one spread across a region. “KP 80% / US 15% / JP 5%” tells you North Korea’s provocation story is being told mostly by Korean press. “US 35% / RU 20% / CN 18%” tells you it’s a great-power story. Same probability, different interpretive weight.85 countries
Why it matters: a threat where the coverage concentrates in one country reads differently from one spread across a region. “KP 80% / US 15% / JP 5%” tells you North Korea’s provocation story is being told mostly by Korean press. “US 35% / RU 20% / CN 18%” tells you it’s a great-power story. Same probability, different interpretive weight.85 countries
KP 26%US 15%KR 13%CN 9%RU 8%+ others 29%
Press provenance
Global 31%GB 18%IN 18%US 12%HK 10%
78%↑ +13.4pp4,350 evidence?Evidence countNumber of dispatches that have moved this threat’s probability up or down under the Bayesian update rule. Each piece of evidence is one scored-and-cited dispatch. More evidence → tighter confidence interval → a probability you can trust.
02 · nuclear · Iran
Iran nuclear escalation
The possibility of Trump proposing that China persuade Tehran to reach an agreement with Washington, which could potentially ease tensions and decrease the probability of Iran nuclear escalation.
Trump arrives in Beijing, Chinese Vice President welcomes him at the airport
+0.4σVS 82D MEAN 71%?What this σ number meansIn plain English: is today’s probability unusual compared to recent history, or business-as-usual?
±1σ (typical) — inside the normal range. Baseline noise.
±1σ to ±2σ (elevated) — outside the usual range. Worth a second look.
Beyond ±2σ (critical) — statistically unusual. Something changed.
Technical: σ (lowercase sigma, the standard-deviation symbol — not to be confused with Σ, the summation symbol) = (current probability − 60-day baseline mean) / baseline standard deviation. Baseline excludes the last 7 days so recent movement isn’t self-referenced.
±1σ (typical) — inside the normal range. Baseline noise.
±1σ to ±2σ (elevated) — outside the usual range. Worth a second look.
Beyond ±2σ (critical) — statistically unusual. Something changed.
Technical: σ (lowercase sigma, the standard-deviation symbol — not to be confused with Σ, the summation symbol) = (current probability − 60-day baseline mean) / baseline standard deviation. Baseline excludes the last 7 days so recent movement isn’t self-referenced.
Geographic scope?Geographic scopeWhere the dispatches that move this threat’s probability are set. We extract the country codes each article references (from NER + its source publication’s country field), aggregate them across all {evidence} pieces that have touched this threat, and show the top 5 by share.
Why it matters: a threat where the coverage concentrates in one country reads differently from one spread across a region. “KP 80% / US 15% / JP 5%” tells you North Korea’s provocation story is being told mostly by Korean press. “US 35% / RU 20% / CN 18%” tells you it’s a great-power story. Same probability, different interpretive weight.178 countries
Why it matters: a threat where the coverage concentrates in one country reads differently from one spread across a region. “KP 80% / US 15% / JP 5%” tells you North Korea’s provocation story is being told mostly by Korean press. “US 35% / RU 20% / CN 18%” tells you it’s a great-power story. Same probability, different interpretive weight.178 countries
IR 30%US 25%IL 9%PK 4%CN 3%+ others 29%
Press provenance
Global 39%IN 19%GB 14%US 14%IR 7%
81%↑ +9.9pp99,234 evidence?Evidence countNumber of dispatches that have moved this threat’s probability up or down under the Bayesian update rule. Each piece of evidence is one scored-and-cited dispatch. More evidence → tighter confidence interval → a probability you can trust.
03 · economic
Global recession
The potential for the UK economy to plunge into recession due to further weeks of disruption across the Strait of Hormuz, which could contribute to a global recession.
Labour leadership turmoil to cost Reeves up to £12bn
+0.5σVS 82D MEAN 81%?What this σ number meansIn plain English: is today’s probability unusual compared to recent history, or business-as-usual?
±1σ (typical) — inside the normal range. Baseline noise.
±1σ to ±2σ (elevated) — outside the usual range. Worth a second look.
Beyond ±2σ (critical) — statistically unusual. Something changed.
Technical: σ (lowercase sigma, the standard-deviation symbol — not to be confused with Σ, the summation symbol) = (current probability − 60-day baseline mean) / baseline standard deviation. Baseline excludes the last 7 days so recent movement isn’t self-referenced.
±1σ (typical) — inside the normal range. Baseline noise.
±1σ to ±2σ (elevated) — outside the usual range. Worth a second look.
Beyond ±2σ (critical) — statistically unusual. Something changed.
Technical: σ (lowercase sigma, the standard-deviation symbol — not to be confused with Σ, the summation symbol) = (current probability − 60-day baseline mean) / baseline standard deviation. Baseline excludes the last 7 days so recent movement isn’t self-referenced.
Geographic scope?Geographic scopeWhere the dispatches that move this threat’s probability are set. We extract the country codes each article references (from NER + its source publication’s country field), aggregate them across all {evidence} pieces that have touched this threat, and show the top 5 by share.
Why it matters: a threat where the coverage concentrates in one country reads differently from one spread across a region. “KP 80% / US 15% / JP 5%” tells you North Korea’s provocation story is being told mostly by Korean press. “US 35% / RU 20% / CN 18%” tells you it’s a great-power story. Same probability, different interpretive weight.132 countries
Why it matters: a threat where the coverage concentrates in one country reads differently from one spread across a region. “KP 80% / US 15% / JP 5%” tells you North Korea’s provocation story is being told mostly by Korean press. “US 35% / RU 20% / CN 18%” tells you it’s a great-power story. Same probability, different interpretive weight.132 countries
US 24%IR 19%IL 5%GB 5%CN 5%+ others 42%
Press provenance
Global 46%GB 18%US 17%AU 13%IN 4%
94%↑ +5.5pp7,164 evidence?Evidence countNumber of dispatches that have moved this threat’s probability up or down under the Bayesian update rule. Each piece of evidence is one scored-and-cited dispatch. More evidence → tighter confidence interval → a probability you can trust.
Risk dimensions?How these indices are builtEach dimension (humanitarian, nuclear, conflict, economic, technology) groups 2+ tracked threats of that category. The dimension index is the cross-sectional average probabilityacross its member threats on any given day — a composite read of whether the whole dimension is running hot or cool.
The sparkline reconstructs this daily average by forward-filling each constituent threat’s last known probability and averaging across the category. Categories with only 1 tracked threat (e.g. currently AI governance) are excluded — “average of one” isn’t an index.category-level composite · 14 days trajectory
The sparkline reconstructs this daily average by forward-filling each constituent threat’s last known probability and averaging across the category. Categories with only 1 tracked threat (e.g. currently AI governance) are excluded — “average of one” isn’t an index.category-level composite · 14 days trajectory
Upcoming conferences and webinars
Watchlist?Bayesian threat boardEach threat has a current probabilitythat moves up or down as new evidence arrives. Each dispatch we ingest is scored against every tracked threat; if it raises or lowers the odds of that threat materializing, it nudges the probability using a Bayesian update (prior + evidence → posterior).
The board shows what our model currently believes, backed by the evidence count(how many dispatches contributed) and confidence intervals. Not forecasting — continuous re-weighting of a claim against all available reporting.
The board shows what our model currently believes, backed by the evidence count(how many dispatches contributed) and confidence intervals. Not forecasting — continuous re-weighting of a claim against all available reporting.
Tracking on our Bayesian threat board.
01
Sudan humanitarian collapse
humanitarian · Sudan
90%
· +1.5pp · 7d
5,571 evidence
02
Global food crisis
humanitarian
84%
· +2.8pp · 7d
8,474 evidence
03
Middle East regional war expansion
conflict · Iran · Israel
79%
· -0.3pp · 7d
138,813 evidence
04
AI governance crisis
technology
73%
↑ +6.6pp · 7d
2,137 evidence
05
China-Taiwan military action
conflict · China · Taiwan
41%
· -0.9pp · 7d
16,486 evidence
Top risk countries · todaytop 10 of 60 · events 90d · score · risk
0133.0Critical0234.8Critical0335.5Critical0435.8Critical0537.7Critical0638.8Critical0740.0Critical0842.0Critical0942.6Critical1043.2CriticalActive conflicts10 tracked · hover for brief
Iran war
Critical100
War
Iran warNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 250,017 dispatches · click to openPersian Gulf conflict
Critical100
Maritime
Persian Gulf conflictNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 60,739 dispatches · click to openMiddle East conflict
Critical100
War
Middle East conflictNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 42,329 dispatches · click to openMexico drug cartel violence
Critical100
Civil War
Mexico drug cartel violenceNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 35,481 dispatches · click to openStrait of Hormuz crisis
Critical100
Maritime
Strait of Hormuz crisisNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 33,789 dispatches · click to openWest Asia conflict
Critical100
War
West Asia conflictNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 27,609 dispatches · click to openIsrael-Hamas war
Critical100
Civil War
Israel-Hamas warNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 27,568 dispatches · click to openUS-China conflict
Critical100
Cold War
US-China conflictNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 26,845 dispatches · click to openIndia-Pakistan conflict
Critical100
Proxy
India-Pakistan conflictNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 26,365 dispatches · click to openSomalia conflict
Critical100
Civil War
Somalia conflictNo daily brief generated yet — this conflict becomes eligible as coverage volume climbs. Click to see tracked data.Escalation 100/100 · 24,220 dispatches · click to openGo deeperfour modes of the product
The Map
One globe, three lenses — pressure, flows, narrative. Time-travel the last 30 days.
Knowledge Graph
Click any country, conflict, actor, or event — the graph recenters. Walk the connections.
Research
Ask the corpus. 1.4M structured rows, 220,000 dispatches, every answer cited.
The Other Side
Every brief, reversed — adversary’s frame, local-language sources elevated, nothing sanitized.