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.
Here’s what Jeremy‘s latest version of his Multi-AI Oracle predicts: (Jeremy’s note: It dropped a bit today. I think the conflicting news reporting is confusing the bot.)
Model value:
* Less than 1000: 1%
* Between 1000 and 3000: 5%
* Between 3000 and 5000: 15%
* Between 5000 and 8000: 35%
* Between 8000 and 12000: 25%
* More than 12000: 19%
Here’s how Jeremy’s Multi-AI Oracle says it arrived at these probabilities:
Obtaining outside data.
Checking internet sources, using keywords: Sudan conflict deaths 2025
Results from around the web: {think}
Okay, let’s tackle this query. The user wants current web results on Sudan conflict deaths in 2025, a summary, tables with relevant data, and a list of sources.
First, I’ll check the provided search results. They all mention the Sudan conflict, but the dates range from late 2024 to early 2025. The user specified 2025, so I need to focus on any 2025 data.
Looking at the sources:
– Source [1] (ACLED) from December 2024 reports over 28,700 fatalities by November 2024, including 7,500 civilians. It mentions an estimate of up to 150,000 deaths but states that’s from a 2024 study.
– Source [2] (Science.org) from February 2025 cites the same 28,700 figure but notes that starvation and disease might cause more deaths than direct violence. It also mentions the U.S. accusing RSF of genocide.
– Source [3] (Genocide Watch) from January 2025 claims 61,000 killed in Khartoum alone, with 26,000 direct war deaths. It references LeMonde’s estimate of 150,000 civilian deaths in November 2024.
– Source [4] (CFR) from March 2025 reports 50 civilians killed in Khartoum recently, indicating ongoing conflict into 2025.
– Source [5] (NRC) from April 2025 states over 150,000 deaths and 14 million displaced.
I need to summarize these findings. The main challenge is reconciling the different death tolls. ACLED and Science.org have lower direct conflict deaths but acknowledge higher estimates. Genocide Watch and NRC cite the 150,000 figure, which includes indirect causes.
For tables, I should create one comparing the sources and another time series if possible. The time series is tricky because the data points are from different months. Maybe group by year or quarter.
Wait, the user asked for a time series with period, value, change, mean, etc. Let’s see: the data spans from late 2024 to early 2025. Let’s structure it quarterly. For example, Q4 2024 with 28,700 deaths, Q1 2025 with 50 new deaths in March. But that’s sparse. Alternatively, monthly? Source [4] mentions March 2025 with 50 deaths. Maybe the time series isn’t extensive, but I’ll present what’s available.
I need to list all the URLs used. The sources provided are [1] to [5], and I should include their links.
Make sure the summary highlights the variance in estimates, the causes (direct vs. indirect deaths), and the humanitarian crisis. The tables should clearly present the data from each source and the time series with statistical measures. Even if the time series data is limited, include it as per the user’s request.
The Sudan conflict has resulted in significant human casualties, with varying estimates depending on the source and methodology. Below is a summary of key findings from recent reports, followed by structured data tables.
—
### Key Findings
– **Reported fatalities** range from **28,700** (direct conflict deaths) to **150,000** (including indirect causes like starvation and disease)[1][2][3][5].
– **Civilian deaths** are estimated at **7,500–26,000** in direct attacks, with Khartoum alone recording over **61,000 total deaths** by late 2024[1][3].
– **Humanitarian impact**: Over **14 million displaced** (30% of Sudan’s population), with both the Sudanese Armed Forces (SAF) and Rapid Support Forces (RSF) accused of blocking aid and using starvation as a weapon[1][2][5].
– **2025 escalation**: March 2025 saw **50 civilians killed** in Khartoum amid intensified fighting[4].
—
### Data Tables
#### **Reported Deaths by Source**
#### **Time Series (Selected Periods)**
—
### URLs Cited
1. [ACLED Conflict Watchlist 2025: Sudan] (https://acleddata.com/conflict-watchlist-2025/sudan/)
2. [Science.org: Sudan Death Toll Analysis] (https://www.science.org/content/article/how-many-have-died-sudan-s-civil-war-satellite-images-and-models-offer-clues)
3. [Genocide Watch: Sudan Genocide Emergency] (https://www.genocidewatch.com/single-post/sudan-genocide-emergency-january-2025)
4. [CFR Global Conflict Tracker] (https://www.cfr.org/global-conflict-tracker/conflict/power-struggle-sudan)
5. [NRC: Sudan Humanitarian Crisis] (https://www.nrc.no/feature/2025/the-humanitarian-crisis-in-sudan-explained)
—
The disparity in death tolls reflects differences in data collection methods and the inclusion of indirect causes. While direct conflict fatalities are better documented, indirect deaths from starvation, disease, and displacement remain underreported[2][3][5]. The conflict has been marked by systemic human rights abuses, including genocide allegations against the RSF[2][3].
OpenAI gives a base rate of the probability that total deaths in the Sudan conflict in 2025 will be Between 5000 and 8000
The question difficulty is rated 7 (0 to 10)
Historical weighted factors include:
Average annual conflict deaths in Sudan over past conflicts, 0.6
Current conflict dynamics and escalation, 0.3
International intervention and humanitarian aid, 0.1
A Bayesian calculation could be performed as follows:
Using the historical factors weightings, current dynamics suggest a higher escalation, likely increasing the base rate from historical averages. Calculated by adjusting 2024 data by estimated increase: 0.3 (base) + 0.15 (escalation) = 0.45.
Bayesian base rate: 0.45
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: 245
The following were considered in order to produce this base rate:
The base rate considers historical averages of conflict deaths in Sudan and the significant escalation observed in late 2024 and early 2025. The deaths range between 5,000 and 8,000 has been chosen due to ongoing conflict and international interventions that may limit further death toll increases.
Ideally, the news feed would contain the following sorts of information for a better forecast:
More precise civilian death numbers and the impact of international humanitarian responses would greatly aid in refining the prediction. Insight into the peace negotiation progress would be valuable.
Some potential divergent considerations that might affect the base rate:
Potential divergent considerations include intensified international military interventions that could either stabilize the situation or cause an increase in violence, and unforeseen escalation factors like new military alliances or breakdowns.
The following chain of events are necessary for the question to resolve positively:
Resolution of the conflict in Sudan or verification of causalities Very unlikely
Querying Claude (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.05, “Between 3000 and 5000”: 0.15, “Between 5000 and 8000”: 0.35, “Between 8000 and 12000”: 0.25, “More than 12000”: 0.19] – confidence: 5)
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”: 0.54, “Between 8000 and 12000”: 0.25, “More than 12000”: 0.05] – 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”: 0.3, “Between 8000 and 12000”: 0.2, “More than 12000”: 0.29] – confidence: 6)
Question Type: Multiple Choice
Confidence: 6
MAPD: Avg: 0.064444444444444, Max: 0.16
# LLM responses: 3
Model value:
* Less than 1000: 1%
* Between 1000 and 3000: 5%
* Between 3000 and 5000: 15%
* Between 5000 and 8000: 35%
* Between 8000 and 12000: 25%
* More than 12000: 19%
The current conflict in Sudan, marked by high casualty rates, saw approximately 28,700 fatalities by November 2024, with continued violence into 2025. The projected number of deaths in the 5,000 to 8,000 range is based on historical conflict data and recent trends, though there is potential for the toll to exceed 8,000 due to persistent hostilities and humanitarian crises, such as the displacement of over 14 million people. Several factors could impact these estimates, including successful peace negotiations or international intervention reducing violence, difficulties in data collection, and new developments in the conflict’s dynamics. While some sources report a broader range of deaths including indirect causes, these figures underscore the ongoing severity and complexity of the situation in Sudan.
Past forecasts on Sudan’s conflict fatalities —>
Guide to Multi-AI Oracle’s Reports (some of these only apply to older versions)
Bayesian calculation: a mathematical technique that helps determine the conditional probability of an event based on prior knowledge and new evidence.
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
Below, map of control in Sudan’s civil war as of April 15, 2025.
Below, before and after pictures of the devastation of the civil war in Sudan.
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/
