TLP:CLEAR · Disclosure is not limited.

Myanmar: civilian casualties and the disputed 100,000-death estimate

Time window: Last 1 day · Audience: General analyst · Type: Situation report · DTG: 2026-11-07 10:09Z · Overall confidence: MEDIUM

BLUF

New reporting attributes an estimate of about 100,000 war deaths to the UN, but the supplied material does not independently verify that figure: its only attribution is carried by a social-media item, and AAPP reports a narrower count of 8,658 people killed by the junta and pro-military groups. Separate reporting documents recent junta air attacks that killed civilians in Sagaing and Karen State, reinforcing the assessment that civilian exposure to air operations remains acute.

Executive summary

Reporting available on 6 and 7 November describes a UN estimate of about 100,000 war deaths and more than 3.5 million people displaced, but the estimate is relayed through social media rather than a supplied UN document. AAPP reports 8,658 people killed by the junta and pro-military groups since 1 February 2021, a differently scoped figure. AAPP also reports 76 verified deaths in Sagaing Region from 1 August to 6 November, 69 attributed to junta airstrikes, and at least five people killed in the 4 November bombing of a public hospital in Kawlin Township. Major-media reporting says an air and drone attack on a Bible school in Day Bu Noh, Hpapun District, on 3 November killed three students and injured six. Reports of strikes in Rakhine State add to the picture but rely on less reliable, social-media-based reporting.

Change from previous assessment

Compared with the 6 November brief, the supplied material now includes a report attributing an estimate of about 100,000 war deaths to the UN. This changes the figure from absent from the supplied evidence to reported, but does not verify it because no primary UN source is provided. New reporting adds AAPP's Sagaing casualty breakdown and accounts of the 3 November Bible-school attack and 4 November hospital bombing, strengthening the basis for assessing ongoing civilian exposure to air operations. The earlier Magway and Thaw Le Hta findings are not updated by this material and are not carried forward as current judgments.

Key judgments

1. The reported UN estimate of about 100,000 deaths in Myanmar's war is not independently substantiated by the supplied material. The estimate is relayed through a social-media item, while AAPP reports 8,658 people killed by the junta and pro-military groups since 1 February 2021; the two figures have different stated scopes and do not directly conflict. (Confidence: low · ASSESSED)
   • I&W: Confirm the estimate's attribution: a primary UN publication or statement gives the approximately 100,000 figure, the period covered and its counting method. (0-14 days)
   • I&W: Break the attribution: a UN statement or correction says the figure was not a UN estimate or provides a materially different total. (0-14 days)
2. Junta air operations are likely to remain a direct civilian hazard over the next 1-3 months. AAPP reports that airstrikes killed 69 of 76 verified victims in Sagaing Region between 1 August and 6 November, and that a 4 November bombing of the public hospital in Shaw Hpyu Kone Village, Kawlin Township, killed at least five people. Major-media reporting says an air and drone attack on a Bible school in Day Bu Noh, Hpapun District, on 3 November killed three students and injured six. Separate reports of 1 November strikes in Rakhine State, including near a temporary clinic and maternity ward in Mrauk-U, carry lower confidence because they are relayed through social media. (Confidence: medium · ASSESSED)
   • I&W: Confirm: AAPP's next dated Sagaing update records new verified civilian deaths from junta airstrikes. (0-14 days)
   • I&W: Break: AAPP's next dated Sagaing update records no new verified airstrike deaths and identifies another cause as the leading cause of death. (0-14 days)
3. Further local displacement in southern Kani Township is likely over the next 0-14 days if the reported junta advance continues. An NGO report says an operation launched from the Northwestern Regional Military Command headquarters in Monywa displaced around 3,000 residents after forces advanced from Chaung Ma village on 4 November. The same reporting stream describes a column of hundreds of troops advancing from Theinzeik through Wadatkwin and Zee Byone to the area of Kaw Khat village by 5 November. These movements are not independently confirmed in the supplied material. (Confidence: low · ASSESSED)
   • I&W: Confirm: dated local reporting describes new evacuations from southern Kani as junta forces continue advancing from Chaung Ma. (0-14 days)
   • I&W: Break: dated local reporting says the Kani column has withdrawn and displaced residents have returned to their villages. (0-14 days)
4. The supplied reporting establishes Thai official concern about military operations near the border, but does not report that the cited incidents harmed people on Thai territory. Thailand's Defence Chief, General Ukrit Boontanon, warned Myanmar's military authorities on 5 November to prevent operations affecting Thai people; a separate report says a 25 October airstrike on Mae Nu Hta village on the Myanmar side killed a child and injured at least two people. This assessment rests on limited reporting from separate sources. (Confidence: low · REPORTED)
   • I&W: Confirm continued official concern: a Thai authority issues a further public warning about Myanmar military operations near the border. (0-14 days)
   • I&W: Break the current no-impact finding: a Thai authority reports harm to people or property inside Thailand from a Myanmar military operation. (0-14 days)

Outlook & scenarios

Continued operations and civilian casualties (65%)

This is the most likely course. Junta air operations and ground advances continue in areas including Sagaing Region, Karen State and Mon State. AAPP records further airstrike deaths in Sagaing, and local reporting describes additional displacement around southern Kani. The national death estimate remains difficult to reconcile with AAPP's narrower count.

Wider strike activity (22%)

Junta air operations expand or intensify across several fronts, with further reported casualties at civilian-use sites in Sagaing, Karen State or Rakhine State. The immediate warning signs are new dated strike reports and updated casualty figures from AAPP, UNICEF or local reporting. Rakhine casualty estimates remain especially vulnerable to source limitations.

Localised lull in reported attacks (10%)

Ground columns pause or withdraw in some areas, and the next reporting period records fewer new civilian casualties or evacuations. This would not establish a nationwide reduction in military operations; evidence would need to show sustained changes across the reported fronts.

Cross-border incident affecting Thailand, low-probability wildcard (3%)

A Myanmar military operation near the Salween River affects people or property in Thailand, prompting a sharper official Thai response. The potential impact would be high, but the supplied reporting documents a Thai warning and a deadly strike on the Myanmar side, not an incident inside Thailand.

Recommendations

1. Verify the approximately 100,000 death estimate against a primary UN publication or statement. Record its time period, definition of war-related deaths, geographic coverage and method before using it as a headline figure.
2. Keep AAPP's count of 8,658 deaths attributed to the junta and pro-military groups separate from any broader war-death estimate. Record the scope and verification status alongside each figure.
3. Maintain a dated incident log for the reported hospital, school and religious-site attacks. Track location, reported aircraft or weapon, casualties, source and whether a second source independently confirms each detail.
4. Seek follow-up on displacement in southern Kani and the reported troop advance from Theinzeik towards Kaw Khat. Distinguish new displacement from residents returning or moving between villages.
5. Monitor statements from Thailand's Defence Chief and other Thai authorities for evidence of harm inside Thai territory. Treat public warnings as indicators of concern, not as confirmation of a cross-border impact.

Confidence & uncertainty

Confidence is medium because reporting from AAPP, major media and NGO sources describes specific civilian casualties and operations across several locations, giving some corroboration for the pattern of continued civilian exposure. Confidence in individual incidents varies, and some Rakhine claims are relayed through social media. The approximately 100,000 death estimate is second-hand in the supplied material and lacks a primary UN source; AAPP's count has a different stated scope.

Intelligence gaps

• [EEI 1.2 · PARTIAL] Changes in the weekly number of civilian deaths, injuries, detentions, enforced disappearances, and reported attacks compared with the preceding four weeks. Recommended collection: UN reporting and incident-monitoring databases
• [EEI 1.3 · PARTIAL] Junta force movements, aircraft sorties, artillery deployments, and preparations near densely populated areas or displacement concentrations. Recommended collection: commercial satellite imagery, local reporting, and military-source reporting
• [EEI 2.1 · PARTIAL] Newly displaced civilians by state or region, including the number of people fleeing each major attack area and the direction of movement. Recommended collection: UN OCHA, humanitarian partners, and local civil-society reporting
• [EEI 2.2 · UNCOVERED] Access conditions and stock levels for food, medicine, fuel, shelter, and clean water in conflict-affected townships. Recommended collection: humanitarian field reporting and satellite imagery
• [EEI 2.4 · PARTIAL] Changes in cross-border refugee flows and requests for emergency assistance in neighboring countries. Recommended collection: border authorities, UNHCR, and humanitarian reporting
• [EEI 3.1 · PARTIAL] Resistance attacks, territorial gains, loss of key transport routes, and changes in control of towns or administrative centers. Recommended collection: conflict-event databases, local media, and geospatial analysis
• [EEI 3.2 · PARTIAL] Junta retaliatory operations, including collective-punishment measures, village burnings, airstrikes, artillery barrages, and arrests following resistance activity. Recommended collection: local human-rights reporting and incident databases
• [EEI 3.3 · UNCOVERED] Evidence of resistance forces establishing evacuation routes, civilian-protection arrangements, local administration, or restrictions on attacks affecting civilian areas. Recommended collection: local governance reporting, NGO interviews, and resistance statements
• [EEI 3.4 · UNCOVERED] Changes in resistance access to weapons, ammunition, drones, communications equipment, training, and cross-border logistical support. Recommended collection: weapons-tracking reporting, border reporting, and imagery analysis
• [EEI 4.1 · PARTIAL] New arms transfers, aviation fuel deliveries, spare-parts shipments, financial flows, or maintenance support reaching the junta. Recommended collection: customs data, sanctions investigations, shipping databases, and commercial imagery
• [EEI 4.2 · PARTIAL] New or expanded sanctions, arms embargo measures, accountability proceedings, diplomatic initiatives, or humanitarian funding commitments. Recommended collection: government, UN, and international-organization statements
• [EEI 4.3 · PARTIAL] Changes in neighboring states' border controls, refugee policies, cross-border aid access, and engagement with junta or resistance representatives. Recommended collection: official statements, border reporting, and diplomatic reporting
• [EEI 4.4 · UNCOVERED] Changes in commercial aviation-fuel prices, war-risk insurance, shipping access, or banking restrictions affecting junta-controlled ports and airports. Recommended collection: maritime/shipping data, financial reporting, and sanctions-monitoring sources

Red cell review: CONCUR WITH COMMENT

TLP:CLEAR