How many conflict deaths will there be in Sudan in 2025?
Jeremy’s most recent version of his Multi-AI Oracle is a side competitor with the machine forecasting bots of the VIEWS competition.
Obtaining outside data.
Checking internet sources, using keywords: Sudan conflict 2025
Results from around the web:
{think}
Okay, let me start by understanding the user’s query. They want current web results related to Sudan conflict deaths in 2025, including data tables and time series if possible. The search results provided are from various sources up to April 2025.
First, I’ll go through each search result to extract relevant death toll figures and dates. Result [1] from ACLED mentions 28,700 reported fatalities by November 2024, including 7,500 civilians. Result [2] from Science.org discusses different estimates, including ACLED’s 28,700 and a model suggesting 61,000 deaths in Khartoum alone. Result [4] from Genocide Watch cites a LeMonde estimate of 150,000 civilian deaths by November 2024. The CFR report in [3] mentions the former U.S. envoy’s estimate of 150,000 killed since April 2023.
I need to organize these figures, noting the sources and dates. Some sources, like [4], highlight that actual numbers might be higher due to underreporting. The WHO report in [5] doesn’t provide direct death counts but mentions the humanitarian crisis. Amnesty International’s statement in [6] focuses on atrocity crimes but doesn’t add new death figures.
For the time series, I can list the estimates chronologically. However, the data points are sparse and from different time frames. The ACLED data is up to November 2024, while the LeMonde estimate is November 2024. The March 2025 WHO report adds context on famine and displacement but not direct deaths.
Creating a table would require aligning the periods, values, and sources. Since the data isn’t uniformly spaced, I’ll present it as a summary table. The time series might not be extensive enough for mean or standard deviation calculations, but I can include the available figures with their changes.
I should also note the challenges in accurate reporting, such as undercounting and blocked access for aid workers, as mentioned in [2] and [4]. Including a comparison of different methodologies (ACLED’s direct counts vs. multiplier models) adds depth.
Finally, wrap it up with a summary of the severity and the need for international attention. Make sure all citations are properly included after each relevant sentence.
The Sudan conflict has reached catastrophic levels by April 2025, with widely varying death toll estimates and deteriorating humanitarian conditions. Key findings from current sources include:
## Conflict Death Estimates
—**Key developments in 2025:**
– SAF regained control of strategic cities (Omdurman, Bahri, Obeid) through Iranian drone-assisted offensives[3][7]
– RSF maintains siege of El Fasher (North Darfur) since September 2024[1][9]
– 30.4 million people require humanitarian assistance (50% of population)[5]
## Humanitarian Impact
– **Displacement:** 14.6 million internally displaced, 2.9 million refugees[4][6]
– **Food Security:** Famine confirmed in North Darfur, 17 areas at risk[5]
– **Health Crisis:**
– 65% of Darfur population lacks medical access[4]
– Cholera outbreaks in IDP camps[4][9]
– 145 healthcare facilities destroyed[4]
## Time Series Analysis of Reported Deaths
*Note: Geometric mean unavailable due to incomplete interval data*
## Critical Challenges
1. **Underreporting:** 90% of Khartoum deaths went unrecorded[4]
2. **Methodology Disparities:**
– Direct counts vs 2.3x multiplier models[2]
– Urban vs rural reporting gaps[9]
3. **Access Restrictions:**
– 22 aid workers killed[4]
– SAF blocked Darfur access since October 2024[4]
The conflict has created the world’s largest displacement crisis[3][6], with both sides using starvation tactics[4][9] and systematic sexual violence[6][9]. Despite UNSC resolutions[3][9], international response remains inadequate with only 6.6% of humanitarian needs funded[6].
OpenAI gives a base rate of probability that the number of deaths will be more than 12,000 by the end of 2025
The question difficulty is rated 8 (0 to 10)
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: 260
The following were considered in order to produce this base rate:
The estimates from various sources indicate past annual deaths ranging significantly higher than 12,000. ACLED, LeMonde, and others report high fatality numbers, exacerbated by ongoing humanitarian crises. Given the scale of displacement and healthcare unavailability, indirect deaths are also considerable. This implies a high probability of the death toll exceeding the 12,000 threshold.
Ideally, the news feed would contain the following sorts of information for a better forecast:
More detailed, real-time tracking of civilian deaths, including independent verification mechanisms in conflict zones, would provide a more accurate picture. Data on indirect mortality causes like famine and disease could also refine estimates.
Some potential divergent considerations that might affect the base rate:
Ceasefire or peace negotiations could lower the rate of fatalities. Improved international intervention or aid accessibility might alleviate indirect mortality causes, such as famine and disease, reducing the death toll.
The following chain of events are necessary for the question to resolve positively: (Carolyn’s note: The Multi-AI Oracle is interpreting “positively” as meaning the results will be consistent with its current forecasting of high likelihoods.)
– The conflict in Sudan continues with significant intensity throughout 2024 and into 2025. Likely
– All parties involved in the conflict remain unable or unwilling to reach a ceasefire or peace agreement. Highly likely
– Humanitarian access remains restricted, leading to high civilian casualties, particularly in regions like Darfur. Likely
– Both direct combat-related deaths and indirect deaths (due to famine and lack of medical care) continue to rise. Highly likely
Querying Claude (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.02, “Between 3000 and 5000”: 0.03, “Between 5000 and 8000”: 0.05, “Between 8000 and 12000”: 0.14, “More than 12000”: 0.75] – confidence: 7)
Querying Mistral (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”: 0.19, “More than 12000”: 0.5] – confidence: 6)
Querying OpenAI (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.02, “Between 3000 and 5000”: 0.05, “Between 5000 and 8000”: 0.11, “Between 8000 and 12000”: 0.15, “More than 12000”: 0.66] – confidence: 8)
Question Type: Multiple Choice
Confidence: 7
MAPD: Avg: 0.055555555555556, Max: 0.16666666666667
# LLM responses: 3
Model value:
(Carolyn’s note: Although currently there are reports of over 3000 conflict fatalities in 2025 so far, sometimes these numbers are adjusted downward due to new information. Indeed, according to Cromwell’s Rule of Statistics, one should never forecast 0% or 1%)
* Less than 1000: 1%
* Between 1000 and 3000: 2%
* Between 3000 and 5000: 5%
* Between 5000 and 8000: 11%
* Between 8000 and 12000: 15%
* More than 12000: 66%
The various AIs have highlighted a severe and escalating humanitarian crisis in Sudan, with high death tolls projected through 2025 due to ongoing conflict, restricted humanitarian aid access, and widespread displacement. Key factors contributing to these predictions include Iranian support for military operations, the siege in El Fasher, and famine in North Darfur, with a staggering 30.4 million people requiring assistance. Historical and current estimates from credible sources like ACLED and Le Monde suggest fatalities already exceed 12,000, and conditions are likely worsening. However, there is hope that effective peace negotiations, improved international interventions, or changes in military tactics could mitigate these outcomes. Despite possible underreporting and methodological discrepancies in counting deaths, these remain critical factors to watch, as they could either validate or skew the predicted fatality rates.
Past forecasts on Sudan’s conflict fatalities —>
Guide to Multi-AI Oracle’s Reports (some of these only apply to older versions)
MAPD: Mean Absolute Percentage Deviation calculates the average percentage difference between actual values and predicted values, providing a relative measure of error.
Median: We collect the output values from the LLMs in an array. We then take the median value. This is between 0 and 1 (i.e. 0.15 = 15%).
Base rate: Currently we take OpenAI’s value as the base rate. We are working on a better value using a different prompt that takes into account process. The base rate helps to determine if the median makes sense.
SD: The standard deviation between the median and the base rate.
Confidence: We query each of the LLMs on how confident they are of their predictions (between 0 and 10) and take a median. Because the LLMs tend to be overconfident, we take anything lower than 6 as being low confidence. This factors into the overall model.
Conf Mode: Based on the confidence value. >=9 is high confidence. Below 6 is low confidence (this is also triggered by an exceptionally high SD).
Mellers: This refers to Barbara Mellers, specifically a paper she wrote that includes a formula for moving values towards an extreme (i.e. 0 or 1).
Reverse Mellers: This uses the formula from above, but with a sub-1 coefficient to move the values closer to 50%.
Theory of Mind: We ask the LLMs what they think other LLMs would predict. We hope that this makes them consider the questions more deeply.
Beta Distribution: Currently unused, but possibly of interest. This is based on the median, the base rate and the SD.
Close Type: We noticed that the appropriate base case for some questions is closer to the extremes, while others are closer to 50%. When we have a low confidence value, this helps us to determine whether to extremize or de-extremize the value. ‘A’ implies closer to zero. ‘B’ implies closer to 50%. ‘C’ implies closer to 100%.
# LLM Responses: The count of LLMs that responded. We query 5 currently, but they have a relatively high API failure rate.
Model value: For normal confidence mode, this is the median. For high confidence, this is the Mellers value. For low confidence, this is either the Mellers or Revers
Background on VIEWS: Near the end of 2024, the latest forecasting competition of the Violence and Impacts Early-Warning System (VIEWS) was launched. On Dec. 4, 2024, during a telecon, a group of machine forecasting experts involved in VIEWS observed Jeremy’s demonstration of his Multi-AI Oracle. Although it was too late for Jeremy to sign up as a VIEWS competitor, they urged him to run a side competition for everyone’s benefit. Jeremy chose their Sudan conflict deaths question.
Background on the Sudan conflict, as evaluated by the team of Jeremy Lichtman and Carolyn Meinel:
(1) The importance of the Sahel, a region in Africa that includes Sudan:
Why the Sahel Actually Matters (Possibly To You, Too)
(2) April 29, 2024: Today, a Russian diplomat is negotiating gold mining with the Sudanese Minister of Minerals.
(3) June 3, 2024: Today the African Union Announced Plans to Mediate in Sudan’s Civil War
(4) Sept. 11, 2024 Over 100K starvation deaths 96% likely in Sudan by 1 Jan 2025
Carolyn has checked Sudan’s news sites below to ensure that they are still in operation. However, given the widespread violence in Sudan, any of these could go offline without warning.
https://www.alnilin.com/
https://www.alrakoba.net/
https://www.altaghyeer.info/ar/
https://www.altaghyeer.info/en/
https://www.alttahrer.com/
https://www.alzaeemsd.com/
https://www.assayha.net
https://www.cover-sd.com/
https://kushnews.net/
https://rakobanews.com
https://www.sudanakhbar.com/
https://sudanjem.com/
https://suna-sd.net/en
https://sudantimes.net/
https://www.sudantribune.net/
https://www.sudantribune.com/
https://sudaneseonline.com/
https://www.sudanile.com/
https://sudanow-magazine.net/
https://www.sudaress.com/
https://www.tagpress.net/
Below, a screenshot from VIEWS forecasting conflict fatalities in Sudan through January 2028. Below that, satellite images of the devastation in Sudan.
Below, before and after pictures of the devastation of the civil war in Sudan.
