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, I need to answer the query about current web results related to Sudan conflict deaths in 2025. Let me start by reviewing the provided search results to extract relevant information.
First, looking at the sources, most of them are from 2024 and early 2025. The user is asking specifically for 2025 data, but the latest information available might be up to April 2025. The key sources here are [1], [2], [3], [4], [5], [6], [7], [9], and [10]. Source [8] seems unrelated as it’s about AI benchmarks, so I’ll ignore that.
From [1], ACLED reports over 28,700 fatalities by November 2024, including 7,500 civilians. They mention that the death toll is likely underestimated, with some estimates as high as 150,000. This source also discusses the conflict’s expansion into Darfur and al-Jazirah, and foreign involvement from countries like Russia and Egypt.
Source [2] from Science.org in February 2025 states that ACLED’s December 2024 report estimated 28,700 deaths from intentional injuries, with another study suggesting 61,000 deaths in Khartoum alone from all causes. They also mention a multiplier effect where indirect deaths (from famine, disease) could be 2.3 times direct deaths, leading to an estimate of 62,000 total deaths by October 2024.
[3] from Genocide Watch in January 2025 reports over 61,000 killed in Khartoum, 26,000 directly by the war, with 90% of deaths unrecorded. They cite a LeMonde estimate of 150,000 civilian deaths by November 2024. This source also discusses ethnic targeting of the Masalit people.
Source [4], CFR’s Global Conflict Tracker from March 2025, mentions a U.S. envoy’s estimate of up to 150,000 deaths since the conflict began in 2023. It highlights over 14 million displaced and 30 million needing aid. Recent fighting in Khartoum in March 2025 resulted in 50 civilian deaths.
[5] from Amnesty International in April 2025 reiterates the 28,700 death toll as of November 2024, with famine conditions and 12.7 million displaced. They note ongoing atrocities and foreign weapons supplies.
[6], Human Rights Watch from April 2025, states tens of thousands killed, half the population facing hunger, and recent SAF advances in Khartoum leading to extrajudicial killings. They mention a March 2025 airstrike in Tora killing dozens.
[7] from Simple Wikipedia (as of April 2025) lists 12,000+ killed and 33,000+ injured by December 2023, but this seems outdated compared to other sources. Probably less reliable due to its simplicity and possible lack of updates.
[9], WHO’s report from March 2025, doesn’t provide direct death counts but mentions 30.4 million needing assistance and famine in North Darfur. Malnutrition and disease outbreaks are significant factors.
[10], a UN Fact-Finding Mission in April 2025, condemns recent massacres in Darfur, with over 100 killed in displacement camps. They note investigations into reprisal attacks by SAF and RSF, and mention 24.6 million facing acute food insecurity.
Now, compiling the data into a time series is challenging because the sources don’t provide monthly breakdowns for 2025. Most data is cumulative up to late 2024 or early 2025. However, I can structure the available data as follows:
– **By November 2024**: 28,700 reported fatalities (ACLED [1][2][5])
– **October 2024**: 14 million displaced (UN via [1])
– **March 2025**: 50 civilians killed in Khartoum (CFR [4])
– **April 2025**: Over 100 killed in Darfur camps (UN [10])
The highest estimate mentioned is 150,000 total deaths (LeMonde via [3], U.S. envoy via [4]). The models in [2] suggest higher numbers when considering indirect deaths.
For tables, I’ll create one summarizing key statistics and another for time series where possible. Since detailed monthly data isn’t available, I’ll note the periods with available figures.
I need to ensure all citations are correctly applied to each data point. Also, clarify that 2025-specific data is limited, with most statistics covering up to early 2025. Highlight the difficulty in accurate reporting due to underrecording and blocked access.
The Sudan conflict, ongoing since April 2023, remains one of the deadliest crises globally, with widely varying death toll estimates due to underreporting and blocked humanitarian access. Below is a summary of current data and analyses:
—
### Key Statistics (2023–2025)
—
### Time Series of Reported Deaths (2024–2025)
—
### Analysis of Discrepancies
– **Underreporting**: Over 90% of deaths in Khartoum go unrecorded [3], and aid blockages hinder data collection [2][6].
– **Indirect deaths**: Malnutrition, disease, and starvation account for most fatalities. For every direct death, models estimate 2.3–15 indirect deaths [2][9].
– **Ethnic targeting**: The RSF has systematically attacked non-Arab groups (e.g., Masalit), with 10,000–15,000 killed in El Geneina (2023) [3][4].
—
### Recent Escalations (2025)
– **SAF advances**: Retook Khartoum’s presidential palace and key cities (Omdurman, Bahri) but faced allegations of reprisal killings [4][6][10].
– **Darfur siege**: RSF attacks on Zamzam and Abu Shouk camps exacerbated famine conditions, with 70+ children killed in El Fasher since January [6][9][10].
– **Foreign involvement**: Iran and Russia supplied drones/fighter jets to SAF, while UAE backed RSF [1][4][5].
—
### Projections
Experts warn the conflict could intensify, with fragmented militias proliferating and starvation worsening. Peace talks remain stalled as both sides reject ceasefires without territorial concessions [1][4][6].
OpenAI gives a base rate of probability range is estimated at the scale provided
(Carolyn’s note: OpenAI failed to return a value, citing below widely varying numbers provided in news reports.)
The question difficulty is rated 8 (0 to 10)
Sufficient news to provide a good forecast? 0 (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 base rate is based on historical data provided up to 2025, with estimates varying widely due to underreporting and indirect effects, such as famine and disease. The most current reliable figures range from 28,700 direct conflict deaths to projections as high as 150,000 total including indirect causes. Given the ongoing conflict, the situation remains fluid.
Ideally, the news feed would contain the following sorts of information for a better forecast:
Access to more granular data, such as monthly summarized death reports specifically for 2025, and direct information from field operations or humanitarian assessments would improve forecast accuracy. Improved transparency and reporting mechanisms from within conflict zones would reduce estimation ambiguity.
Some potential divergent considerations that might affect the base rate:
Resolutions greater than 12,000 deaths for 2025 explicitly depend on continued severe violence and limited international intervention. Fluctuations in violence levels or successful peace interventions could adjust these projections downward.
The following chain of events are necessary for the question to resolve positively:
– Continuing conflict in Sudan leading to civilian casualties. Almost certain
– Limited access for humanitarian agencies causing increased indirect deaths. Very likely
– Fragmentation of militias leading to disparate fighting across regions. Likely
– Failure of peace talks maintaining ongoing violence. Very likely
– Major offensives by SAF or RSF resulting in further casualties. Likely
– Underreporting and inaccessibility leading to underestimated death tolls. Almost certain
Querying Claude (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.02, “Between 3000 and 5000”: 0.07, “Between 5000 and 8000”: 0.15, “Between 8000 and 12000”: 0.25, “More than 12000”: 0.5] – confidence: 6)
Querying Mistral (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.1, “Between 3000 and 5000”: 0.2, “Between 5000 and 8000”: 0.3, “Between 8000 and 12000”: 0.29, “More than 12000”: 0.1] – confidence: 6)
Querying OpenAI (AI predicts: [“Less than 1000”: 0.01, “Between 1000 and 3000”: 0.02, “Between 3000 and 5000”: 0.07, “Between 5000 and 8000”: 0.1, “Between 8000 and 12000”: 0.2, “More than 12000”: 0.6] – confidence: 6)
Question Type: Multiple Choice
Confidence: 6
Mean Absolute Percentage Deviation
MAPD: Avg: 0.11111111111111, Max: 0.33333333333333
# LLM responses: 3
Model value:
* Less than 1000: 1%
* Between 1000 and 3000: 2%
* Between 3000 and 5000: 7%
* Between 5000 and 8000: 15%
* Between 8000 and 12000: 25%
* More than 12000: 50%
The ongoing conflict in Sudan has led to significant casualties, with direct deaths recorded at over 28,700 and total estimated deaths reaching up to 150,000 by the end of 2024, including those from indirect causes such as famine and disease. Escalations in early 2025, along with systemic underreporting, limited humanitarian access, failed peace talks, and expanding violence, suggest a continuation of high death tolls in 2025. However, the possibility of successful peace negotiations, international interventions improving humanitarian aid access, and reducing indirect deaths could lead to lower casualty estimates, though these potential positive developments face challenges in the volatile conflict environment. The uncertainty surrounding underreporting and the impact of indirect deaths adds difficulty to accurate projections. The conflict’s fluid nature and unpredictable international responses mean that predictions on death tolls remain speculative, contingent on future developments.
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.
