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 deaths 2025
Results from around the web: {think}
Okay, I need to answer the user’s query about current web results related to Sudan conflict deaths, including summarizing the contents and creating relevant tables. Let me start by reviewing the search results provided.
First, looking at the sources, there are several key pieces of information. Source [1] from ACLED mentions over 28,700 reported fatalities by November 2024, including 7,500 civilians, but notes that this is an underestimate. Source [7], Genocide Watch, cites LeMonde’s estimate of at least 150,000 civilian deaths by November 2024. Another source, [8] from Wikipedia, states that as of November 2024, at least 61,000 were killed in Khartoum alone, with 26,000 from direct violence. Source [10] discusses different methodologies for estimating deaths, with ACLED’s figure at 28,700 and another model suggesting 61,000 deaths in Khartoum.
I notice discrepancies in the numbers. For example, ACLED’s 28,700 is much lower than Genocide Watch’s 150,000. The user also wants tables. I should categorize the data into direct conflict deaths and total deaths including indirect causes. There’s mention of a death multiplier from past conflicts (source [10]), which can help in estimating total deaths. Studies referenced in [10] suggest multipliers ranging from 2.3 to 15, leading to a wide range of estimates.
Also, [10] mentions that Yale researchers and others use satellite imagery and modeling to estimate casualties, which could be a method to note. The UN report in [4] documents a spike in civilian deaths in February 2025, but the current date is April 2025, so the latest data might be from February.
I need to create a table summarizing these estimates. Let’s list each source, their estimate, time period, and methodology. Another table could show the time series data, but the available information doesn’t provide monthly breakdowns, only cumulative figures up to certain dates. However, the UN’s report in [4] mentions 275 civilian deaths in a week in February 2025, which is a significant spike. I can format that into a time series table.
Potential challenges: Ensuring accuracy given the conflicting sources and different methodologies. Also, some sources might overlap in their data (like ACLED and the Yale study). Need to present them clearly without conflating different figures.
I should note that the actual death toll is likely higher due to underreporting and blocked access for data collection.
The user also asked for tables in CSV or tabbed format. Let me structure the summary first, then the tables.
Include key points: direct vs indirect deaths, sources of data, multipliers, and recent spikes. Also, mention the difficulty in obtaining accurate numbers due to the conflict and blockades by warring parties.
The Sudanese conflict (April 2023–present) has generated highly contested death toll estimates due to limited access for data collection and deliberate obstruction by warring factions. Key findings from current sources:
—
### **Key Estimates**
—
### **Casualty Multipliers** (Indirect vs. Direct Deaths) [10]
### **Time Series: Civilian Deaths in Khartoum** [4][7][8]
—
### **Key Trends**
– **Underreporting crisis**: 90% of Khartoum deaths went unrecorded [7], and conflict zones like Darfur lack systematic data collection[10].
– **Targeted atrocities**: The RSF has killed 10,000–15,000 Masalit people in West Darfur (2023–2024) [7].
– **Humanitarian collapse**: 65% of Darfur lacks healthcare access, and 10 children/week die from malnutrition in Chadian refugee camps[7].
—
### **Methodological Challenges** [10]
1. **Blocked access**: Both SAF and RSF prohibit aid workers from documenting casualties.
2. **Satellite limitations**: Yale researchers use nighttime light reductions (−72% in Khartoum) and crowd-sourced obituaries but lack granular data.
3. **Displacement bias**: 14M+ displaced Sudanese skew mortality models[1][7].
—
This data underscores the conflict’s catastrophic human toll, with credible estimates ranging from 28,700 to over 150,000 deaths when including indirect causes. Recent escalations suggest fatalities will continue rising through 2025[1][8].
OpenAI gives a base rate of (Carolyn’s note the base rate is discussed below but no percentages given because of poor data.)
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: 365 (Carolyn’s note: the bot is confused. There are only 272 days left.)
The following were considered in order to produce this base rate:
The base rate is derived from aggregated data showing significant historical death tolls from similar conflicts, particularly within Sudan’s ongoing conflict context since April 2023. Using the midpoint estimates of between 150,000 total deaths from sources like Genocide Watch and observed trends in direct and indirect casualties, we conclude that the likelihood of the question resolving as any specific category over 12,000 is high given the underreporting and humanitarian collapse.
Ideally, the news feed would contain the following sorts of information for a better forecast:
Access to real-time satellite imagery and more granular updates on mortality from conflict zones would improve confidence in forecasts. Verification from UN bodies or international organizations with on-ground access could also enhance data accuracy.
Some potential divergent considerations that might affect the base rate:
Discrepancies in data estimates may arise due to limited access for data collection and obstruction by warring parties. As such, actual numbers could vary significantly depending on future data reliability and verification efforts.
The following chain of events are necessary for the question to resolve over 12,000:
– Continued armed conflict in Sudan resulting in further casualties,Very likely
– Lack of effective international intervention or peacekeeping Likely
– Persistent humanitarian crisis exacerbating mortality Likely
Querying Claude (Error: Claude is not returning an array)
(Carolyn’s note: Failing to give an answer –‘array’ — is a common problem with even the biggest of the GenAIs)
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”: 0.24, “Between 8000 and 12000”: 0.3, “More than 12000”: 0.3] – 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”: 0.2, “Between 8000 and 12000”: 0.24, “More than 12000”: 0.4] – confidence: 6)
Question Type: Multiple Choice
Confidence: 6
# LLM responses: 2
Model value:
* Less than 1000: 1%
* Between 1000 and 3000: 5%
* Between 3000 and 5000: 10%
* Between 5000 and 8000: 22%
* Between 8000 and 12000: 27%
* More than 12000: 35%
The predictions regarding the death toll in the Sudan conflict are primarily based on historical data and recent estimates, which indicate significant casualties since April 2023. Several sources, including Genocide Watch, estimate the death toll in the tens of thousands. Underreporting and the humanitarian crisis complicate accurate assessment, leading to a high likelihood of deaths exceeding 12,000 by 2025. Discrepancies in estimates are due to limited data access and potential obstruction by warring parties, making actual figures likely to vary. However, the predictions could be challenged by conflicting data, potential international interventions, peacekeeping efforts, or a de-escalation of the conflict, which could lead to a reduction in the anticipated death toll.
Past forecasts on Sudan’s conflict fatalities —>
Guide to Multi-AI Oracle’s Reports (some of these only apply to older versions)
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.
