Botmaster Jeremy Lichtman’s Sudan conflict deaths forecast, June 30, 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: 6%
* Between 3,000 and 5,000: 25%
* Between 5,000 and 8,000: 37%
* Between 8,000 and 12,000: 25%
* More than 12,000: 6%

11–14 minutes read

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

The web content you retrieved contains a report on the conflict in Sudan, specifically regarding the underreporting of deaths due to the ongoing civil war. The report was conducted by researchers at the London School of Hygiene & Tropical Medicine (LSHTM) and highlights the severe impact of the conflict on Sudanese lives.

Key points from the report include:
1. **Underreporting of Deaths**: The study indicates that the actual death toll in the Sudan war is much higher than previously reported. In the first 14 months of the conflict (April 2023 to June 2024), over 61,000 people died of all causes in Khartoum State alone, marking a 50% increase in the pre-war death rate. Of these, 26,000 deaths were directly due to violence, surpassing the 20,178 intentional-injury deaths reported for the entire country by the Armed Conflict Location and Event Data (ACLED).
2. **Methodology and Findings**: The researchers utilized “capture-recapture analysis” to estimate unrecorded deaths by comparing data from multiple sources such as public surveys, social media obituaries, and private networks. They found a significant number of deaths went unrecorded, especially from preventable diseases and starvation. Regions like Kordofan and Darfur had disproportionately high death rates due to violence.
3. **Impact of the Conflict**: The war, which broke out in April 2023 between the Sudanese army and the Rapid Support Forces (RSF), has led to dire humanitarian consequences. There’s widespread starvation, with three major aid organizations warning of a “historic” hunger crisis. Moreover, United Nations experts accused the warring factions of using “starvation tactics” as a weapon, affecting 25 million civilians.
4. **International Response**: Though there are ongoing international efforts to end the violence, including diplomatic interventions and aid, the conflict persists. Russian involvement has complicated ceasefire attempts, with Russia vetoing a UN Security Council resolution calling for a ceasefire.
5. **Humanitarian Situation**: The United Nations reports that the conflict has displaced 11 million people, with half of Sudan’s population requiring food aid, constituting the world’s biggest hunger crisis. Aid groups, such as Doctors Without Borders, have observed a rise in severe malnutrition and serious injuries among Sudanese children.
Overall, the content underscores the devastating effects of the civil war in Sudan and highlights the challenges in accurately assessing and addressing the humanitarian needs due to data limitations and ongoing violence.

Fact-checking news summary:
Here is a list of specific facts from the summary, along with an assessment of their correctness, importance, and relevance to the original question regarding the number of state-based conflict deaths reported by ACLED in Sudan in 2025:
1. **Underreporting of Deaths**:
– **Fact**: Over 61,000 people died of all causes in Khartoum State from April 2023 to June 2024.
– **Correctness**: True (based on the summary information).
– **Importance**: Important – Highlights the extent of underreporting compared to ACLED’s data.
– **Relevance**: True – Provides context for why future ACLED reports may also be understated.
2. **Conflict-Related Deaths**:
– **Fact**: 26,000 deaths in Khartoum State were directly due to violence.
– **Correctness**: True.
– **Importance**: Important – Demonstrates the scale of direct conflict-related violence.
– **Relevance**: True – Directly pertains to conflict deaths that may be relevant to future reporting by ACLED.
3. **ACLED Reporting**:
– **Fact**: ACLED reported 20,178 intentional-injury deaths for the entire country over the same period.
– **Correctness**: True, given the summary’s context.
– **Importance**: Critical – Directly pertains to the difference between reported and actual conflict death figures.
– **Relevance**: True – Crucial for understanding any discrepancy in future reported numbers.
4. **Research Methodology**:
– **Fact**: The researchers used capture-recapture analysis.
– **Correctness**: True.
– **Importance**: Important – Ensures the results are based on a credible and systematic approach.
– **Relevance**: True – Ensures the reliability of underreporting claims which affects future predictions.
5. **Conflict Timeline**:
– **Fact**: The conflict began in April 2023.
– **Correctness**: True.
– **Importance**: Important – Establishes the timeframe for evaluating the conflict’s impact.
– **Relevance**: True – Relevant to understanding trends and potential future developments.
6. **Impact Assessment**:
– **Fact**: Three major aid organizations warned of a “historic” hunger crisis.
– **Correctness**: True.
– **Importance**: Less important – Focused more on humanitarian aid rather than direct conflict deaths.
– **Relevance**: False – Less directly related to the question of conflict deaths.
7. **International Complications**:
– **Fact**: Russia vetoed a UN Security Council resolution calling for a ceasefire.
– **Correctness**: True.
– **Importance**: Less important – Involves international politics more than conflict death statistics.
– **Relevance**: False – Not directly related to conflict death figures.
8. **Human Displacement**:
– **Fact**: The conflict displaced 11 million people.
– **Correctness**: True.
– **Importance**: Important – Demonstrates the scale of the humanitarian crisis.
– **Relevance**: True – Indicates instability which could affect future conflict death reporting.
9. **Malnutrition and Child Impact**:
– **Fact**: Rise in severe malnutrition and serious injuries among Sudanese children.
– **Correctness**: True.
– **Importance**: Important – Related to the broader effects of the conflict but not directly measuring deaths.
– **Relevance**: False – More about humanitarian impact than conflict death statistics.

This analysis helps clarify the current problems in accurately reporting conflict deaths and which factors are likely to affect future ACLED data reporting.

Fact-checked summary:
The conflict in Sudan, which began in April 2023, has resulted in significant underreporting of state-based conflict deaths. From April 2023 to June 2024, over 61,000 people died of various causes in Khartoum State, with 26,000 of those deaths directly attributed to violence. However, ACLED reported only 20,178 intentional-injury deaths for the entire country during the same period, highlighting a discrepancy in reported figures. The underreporting is significant and indicates that future ACLED reports may also understate conflict deaths. This underreporting is supported by the researchers’ use of capture-recapture analysis, ensuring the findings are based on a systematic methodology. Additionally, the conflict has displaced 11 million people, suggesting a level of instability that may further affect the accuracy of future conflict death reports by ACLED.

OpenAI gives a base rate of Between 5000 and 8000
The question difficulty is rated 7 (0 to 10)
Historical weighted factors include:
Underreporting of conflict deaths, 0.4
Lack of access to conflict areas, 0.3
Displacement leading to undercounting, 0.3
A Bayesian calculation could be performed as follows:
P(Between 5000 and 8000 | Historical factors) = (0.4 * P(Underreporting being accounted for correctly) + 0.3 * P(Access improving) + 0.3 * P(Displacement counted correctly)) * Base Rate = 0.2. Assume probabilities of improvements in each factor are low, leading to slight increase in overall probability.
Bayesian base rate: 0.22
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: 730
The following were considered in order to produce this base rate:
The base rate was derived by considering both historical underreporting and the typical range of deaths in similar conflicts. Past conflicts often show official numbers underreport the true toll due to various factors like lack of access and institutional inadequacies.

Ideally, the news feed would contain the following sorts of information for a better forecast:
Access to on-the-ground sources and additional regional reports from non-governmental organizations could provide a more accurate assessment. Monitoring any international negotiations that could lead to conflict resolution would also be helpful.

Some potential divergent considerations that might affect the base rate:
If significant improvements in international attention and methodological changes in data collection are made, reported numbers could diverge upwards. Conversely, ongoing instability and limited access could further depress the numbers reported by ACLED.

The following chain of events are necessary for the question to resolve positively:
– Resolution of the Sudan conflict with comprehensive reporting of conflict deaths from multiple reliable sources Very unlikely
– Significant international intervention to reduce conflict and improve stability in Sudan Unlikely
– Improvement in data collection methodologies and reduction in underreporting Possible

Querying Claude (AI predicts: [“Less than 1000”: 0.05, “Between 1000 and 3000”: 0.15, “Between 3000 and 5000”: 0.25, “Between 5000 and 8000”: 0.3, “Between 8000 and 12000”: 0.2, “More than 12000”: 0.05] – confidence: 6)

Querying Mistral (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.05, “Between 3000 and 5000”: 0.2, “Between 5000 and 8000”: 0.5, “Between 8000 and 12000”: 0.2, “More than 12000”: 0.04] – confidence: 6)

Querying OpenAI (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.05, “Between 3000 and 5000”: 0.15, “Between 5000 and 8000”: 0.25, “Between 8000 and 12000”: 0.29, “More than 12000”: 0.25] – confidence: 6)

Question Type: Multiple Choice
Confidence: 6
MAPD: Avg: 0.087777777777778, Max: 0.16666666666667
# LLM responses: 3
Explanations of the above statistical measures here —>

Model value:
* Less than 1,000: 1%
* Between 1,000 and 3,000: 6%
* Between 3,000 and 5,000: 25%
* Between 5,000 and 8,000: 37%
* Between 8,000 and 12,000: 25%
* More than 12,000: 6%

The AIs predict that the number of conflict-related deaths in Sudan is underreported due to systemic issues with data collection, with ongoing instability and displacement of 11 million people exacerbating the situation. Historical data and current trends suggest a base rate of 5000 to 8000 deaths, though underreporting may obscure the true figure. Key reasons for potential inaccuracies include limited access to conflict zones, possible improvements or changes in data collection, and unforeseen developments such as international intervention or shifts in conflict intensity, which could either increase or decrease the reported death toll.

Runtime: 160 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.