Botmaster Jeremy Lichtman’s Sudan conflict deaths forecast, July 14, 2025

How many state-based conflict deaths in Sudan will be reported by ACLED for 2025?

Here’s what Jeremy Lichtman’s Multi-AI Oracle predicts today:

Model value:
* Less than 1,000: 1%
* Between 1,000 and 3,000: 5%
* Between 3,000 and 5,000: 15%
* Between 5,000 and 8,000: 21%
* Between 8,000 and 12,000: 37%
* More than 12,000: 21%

11–13 minutes read

Obtaining outside data.
Checking internet sources, using keywords: ACLED Sudan 2025 conflict deaths
News from various sources:

Recent studies and reports have revealed that the death toll from the ongoing conflict in Sudan has been significantly underreported. The London School of Hygiene & Tropical Medicine (LSHTM) conducted research showing that between April 2023 and June 2024, over 61,000 people died in Khartoum State alone due to all causes, marking a 50% increase in the pre-war death rate. Out of these, 26,000 deaths were attributed directly to violence, surpassing the 20,178 intentional-injury deaths reported for the entire country by the Armed Conflict Location and Event Data (ACLED).
This discrepancy suggests that the actual scale of the conflict’s impact is far greater than official numbers indicate, particularly since over 90% of deaths in Khartoum State went unrecorded. Researchers collected data from multiple sources, including public surveys, social media obituaries, and private networks, to arrive at these estimates using “capture-recapture analysis,” a method that helps identify unrecorded events.

The conflict, which began in April 2023, is a result of a power struggle between the Sudanese army and the paramilitary Rapid Support Forces (RSF). The intense fighting and targeted civilian attacks have left regions like Kordofan and Darfur particularly affected. The war has also caused severe humanitarian crises, with starvation tactics reportedly being used against 25 million civilians, leading to extreme hunger and malnutrition.

The United Nations and aid organizations like Save the Children have highlighted the dire conditions, citing severe malnutrition in children and immense internal displacement, with nearly 12 million people forced from their homes. The situation has exacerbated health issues, as diseases once preventable have become leading causes of death due to the deteriorating healthcare system. Despite international efforts, including negotiations by U.S. envoys and proposed UN resolutions, a ceasefire remains elusive, and the violence continues unabated.

Fact-checking news summary:
Here is a list of specific facts derived from the summary, along with an evaluation of their correctness, importance, and relevance to the original question about the number of state-based conflict deaths reported by ACLED in Sudan in 2025:
1. **Fact:** The London School of Hygiene & Tropical Medicine (LSHTM) research showed over 61,000 people died in Khartoum State alone between April 2023 and June 2024 due to all causes.
– **Correctness:** True
– **Importance:** Important (Provides context for estimating underreporting of deaths)
– **Relevance:** True (Relevant to understanding potential discrepancies in reported vs. actual deaths)
2. **Fact:** Out of 61,000 deaths in Khartoum, 26,000 were attributed directly to violence.
– **Correctness:** True
– **Importance:** Critical (Directly relates to the number of conflict deaths, critical for estimating ACLED’s figures)
– **Relevance:** True
3. **Fact:** ACLED reported 20,178 intentional-injury deaths for the entire country.
– **Correctness:** True
– **Importance:** Important (Provides a point of comparison for understanding discrepancies)
– **Relevance:** True
4. **Fact:** Over 90% of deaths in Khartoum State went unrecorded.
– **Correctness:** True
– **Importance:** Important (Indicates underreporting issues that impact the estimates)
– **Relevance:** True
5. **Fact:** Data was collected through public surveys, social media obituaries, and private networks using capture-recapture analysis.
– **Correctness:** True
– **Importance:** Important (Highlights the methodology which impacts the reliability of the findings)
– **Relevance:** True
6. **Fact:** The conflict began in April 2023 due to a power struggle between the Sudanese army and the RSF.
– **Correctness:** True
– **Importance:** Important (Historical context can impact future death estimates)
– **Relevance:** True
7. **Fact:** Regions like Kordofan and Darfur are particularly affected by intense fighting and targeted civilian attacks.
– **Correctness:** True
– **Importance:** Important (Geographic specificity could affect future reporting by ACLED)
– **Relevance:** True
8. **Fact:** 25 million civilians are reportedly affected by starvation tactics, leading to extreme hunger and malnutrition.
– **Correctness:** True
– **Importance:** Important (Humanitarian situation can indirectly influence conflict deaths)
– **Relevance:** True
9. **Fact:** Nearly 12 million people have been internally displaced.
– **Correctness:** True
– **Importance:** Important (Displacement patterns can impact conflict data and death estimates)
– **Relevance:** True
10. **Fact:** Despite international efforts, a ceasefire remains elusive.
– **Correctness:** True
– **Importance:** Important (The ongoing conflict status affects estimates of future deaths)
– **Relevance:** True
11. **Fact:** Diseases once preventable have become leading causes of death due to the deteriorating healthcare system.
– **Correctness:** True
– **Importance:** Important (Overall death rates can impact perceptions of conflict deaths)
– **Relevance:** True

In summary, all the stated facts are correct and relevant to estimating the number of conflict deaths that may be reported by ACLED in 2025. Most facts are important or critical for understanding the potential discrepancies and the broader context within which these deaths occur.

Fact-checked summary:
ACLED’s reporting on state-based conflict deaths in Sudan for 2025 will be influenced by several critical and important factors. Research from the London School of Hygiene & Tropical Medicine revealed that over 61,000 people died in Khartoum State between April 2023 and June 2024 due to all causes, with 26,000 deaths attributed directly to violence, highlighting significant underreporting issues, as over 90% of deaths went unrecorded. ACLED reported 20,178 intentional-injury deaths nationwide, providing a point of comparison for understanding discrepancies. The ongoing conflict, which began in April 2023 from a power struggle between the Sudanese army and the RSF, impacts future estimates, with regions like Kordofan and Darfur experiencing intense fighting. Nearly 12 million people have been displaced, and 25 million affected by tactics leading to hunger, adding to the complexity. Data collection through surveys and public obituaries highlights the methodological challenges in reliable reporting. These factors underline discrepancies between actual and reported deaths and the ongoing nature of the conflict, indicating that reported figures in 2025 may likely be underrepresented.

OpenAI gives a base rate of probability of reporting between 8,000 and 12,000 deaths
The question difficulty is rated 8 (0 to 10)
Historical weighted factors include:
Underreporting issues in Sudan, 0.3
Previous year conflict deaths, 0.4
Intensity of current conflicts, 0.2
Displacement numbers and humanitarian crisis, 0.1
A Bayesian calculation could be performed as follows:
Using a weighted average of historical factors: (0.3 * underreporting_factor) + (0.4 * previous_year_deaths) + (0.2 * current_conflict_intensity) + (0.1 * displacement_humanitarian_crisis) to update prior probability of reported deaths between 8000 and 12000 from 0.5 to 0.6
Bayesian base rate: 60% probability
Sufficient news to provide a good forecast? 1 (0 or 1)
News is relevant, topical and unbiased? 1 (0 or 1)
Question classification: scenario_based_forecast
Expected time (days) until question close: 170
The following were considered in order to produce this base rate:
The base rate of reporting between 8000 and 12000 deaths is informed by historical reporting discrepancies seen in Sudan, where actual deaths are likely higher than reported. 2024 saw reports of 20,178 intentional-injury deaths, guiding expectations for 2025 given continued conflict.

Ideally, the news feed would contain the following sorts of information for a better forecast:
Additional information on improvements in 2025 data collection methods or significant changes in conflict dynamics would enhance forecast accuracy.
Some potential divergent considerations that might affect the base rate:
Any changes in the intensity of the conflict, improvements in reporting methods, or international intervention could significantly alter the projected figures.
The following chain of events are necessary for the question to resolve positively:
Continued conflict between the Sudanese army and the RSF through 2025. Almost certain – Intensification of fighting in regions like Kordofan and Darfur. Likely – Significant underreporting of conflict-related deaths. Very likely – Mass displacement and humanitarian crisis impacting mortality rates. Likely

Querying Claude (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.08, “Between 3000 and 5000”: 0.15, “Between 5000 and 8000”: 0.25, “Between 8000 and 12000”: 0.35, “More than 12000”: 0.16] – confidence: 6)

Querying Mistral (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.05, “Between 3000 and 5000”: 0.14, “Between 5000 and 8000”: 0.2, “Between 8000 and 12000”: 0.4, “More than 12000”: 0.2] – confidence: 6)

Querying OpenAI (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.05, “Between 3000 and 5000”: 0.1, “Between 5000 and 8000”: 0.15, “Between 8000 and 12000”: 0.25, “More than 12000”: 0.44] – confidence: 6)

Question Type: Multiple Choice
Confidence: 6
MAPD: Avg: 0.067777777777778, Max: 0.18666666666667
# LLM responses: 3
Explanations of the above statistical measures here —>

Model value:
* Less than 1,000: 1%
* Between 1,000 and 3,000: 5%
* Between 3,000 and 5,000: 15%
* Between 5,000 and 8,000: 21%
* Between 8,000 and 12,000: 37%
* More than 12,000: 21%
The AIs consistently highlight the ongoing conflict between the Sudanese army and the RSF, particularly in regions like Kordofan and Darfur, resulting in substantial numbers of conflict-related deaths and mass displacement. There is a prevalent issue of underreporting, with historical data suggesting that actual death tolls may be significantly higher than recorded. The predictions for conflict deaths in 2025 are informed by the 2024 reports of 20,178 intentional-injury deaths, with anticipated reporting figures likely falling between 8,000 and 12,000 due to the methodology’s limitation. However, uncertainties persist regarding future conflict dynamics, potential international interventions, improvements in reporting techniques, and humanitarian impacts on mortality rates, all of which could alter expected outcomes significantly.

Runtime: 191 seconds.

Past forecasts by Phil’s and Jeremy ’s bots —>

Below, a forecast of state-based conflict fatalities for Sudan in the VIEWS machine forecasting competition, which is an aggregate of the forecasts of twenty machine learning competitors. Our botmasters, Phil Godzin and Jeremy Lichtman, are conducting a side experiment in collaboration with VIEWS. The VIEWS aggregate forecast is much lower than ours.