How many state-based conflict deaths (total of all civilian and combat deaths, including both Ukrainian and Russian combatants) will be reported by ACLED in Ukraine in September 2025?
Here’s what Jeremy’s Multi-AI Oracle predicts today:
Less than 500: 10%
Between 500 and 1,000: 20%
Between 1,000 and 1,500: 30%
Between 1,500 and 2,000: 25%
Greater than 2,000: 15%
11–14 minutes read
Obtaining outside data.
Checking internet sources, using keywords: Ukraine conflict deaths September 2025
News from various sources:
The web content offers a detailed overview of the ongoing conflict in Ukraine as of mid-September 2025, focusing on various military, political, and humanitarian aspects. Below is a summary of the key points related to the Ukraine conflict and reported deaths up to September 2025:
1. **Russian Combat Losses**: The reported total combat losses for the Russian troops are approximately 1,095,520 individuals since the full-scale invasion began on February 24, 2022. Recently, 910 soldiers have been added to the death toll over a single day. Additionally, significant Russian military equipment losses include tanks, armored vehicles, artillery systems, and more.
2. **Recent Offensive and Defensive Operations**:
– Russia recently conducted a large-scale attack involving nearly 540 drones and 45 missiles across 14 Ukrainian regions. This resulted in at least one death and 30 injured.
– Meanwhile, Ukrainian forces have been effective in defending against such attacks, neutralizing 48 of 75 Russian drones in a recent engagement.
3. **Political and Diplomatic Developments**:
– Efforts from global leaders, including former U.S. President Donald Trump, continue in seeking diplomatic solutions and exerting pressure on Russia. Trump has previously indicated he could end the conflict swiftly but has become more pessimistic about an immediate resolution.
– The United Nations Secretary-General has condemned Russian attacks, particularly a strike on government buildings in Kyiv, labeling them as escalations and calling for a ceasefire.
4. **Military Cooperation and Support**:
– The U.S. is providing ongoing military assistance to Ukraine, including support for Patriot Air Defense Systems and delivering armored vehicles.
– Financial and military support also continues from EU and Western countries to bolster Ukraine’s defense capabilities against Russian aggression.
5. **Human Impact and Casualties**:
– UN and independent assessments highlight continued civilian casualties, including the deaths of children, due to Russian strikes.
– There are ongoing investigations into potential corrupt practices in military supply contracts within Ukraine, aimed at ensuring resources are properly utilized.
6. **Future Prospects**:
– Tensions remain high as joint military exercises by Russia and Belarus, dubbed “Zapad 2025,” have raised alarms in Eastern Europe.
– The conflict continues to affect global geopolitical dynamics, with discussions around sanctions on Russia and the strategic positioning of Western allies.
The situation is continually evolving with diplomatic, military, and humanitarian implications.
Fact-checking news summary:
Below is a list of specific facts from the summary, each evaluated for correctness, importance to the resolution of the question, and relevance to the outcome:
1. **Fact**: The reported total combat losses for Russian troops are approximately 1,095,520 individuals since the invasion began on February 24, 2022.
– **Correctness**: This number is implausibly high for the time frame stated. NAT0 reported estimates likely to be considerably lower than this.
– **Importance**: Critical – It directly relates to calculating the number of conflict deaths to address the original question.
– **Relevance**: True – It is relevant since it pertains to the total number of deaths in the conflict.
2. **Fact**: Recently, 910 Russian soldiers have been added to the death toll over a single day.
– **Correctness**: Potentially inflated, such daily losses exceed typical reporting thresholds unless confirmed by independent sources.
– **Importance**: Important – It provides context on the intensity of the conflict.
– **Relevance**: True – Relevant to ongoing conflict death estimates.
3. **Fact**: Russia conducted a large-scale attack involving 540 drones and 45 missiles across 14 Ukrainian regions, resulting in at least one death and 30 injured.
– **Correctness**: Generally plausible, though specific figures vary based on reports.
– **Importance**: Lesser – While demonstrating the conflict’s scale, it is less critical for the overall death count.
– **Relevance**: True – Relevance to overall context and understanding of conflict operations.
4. **Fact**: Ukrainian forces neutralized 48 of 75 Russian drones in a recent engagement.
– **Correctness**: Plausible within the context of recent military engagements.
– **Importance**: Lesser – Details specific engagements but not the total death toll.
– **Relevance**: True – Provides context on military balance and engagement dynamics.
5. **Fact**: Former U.S. President Donald Trump has expressed more pessimism about an immediate resolution to the conflict.
– **Correctness**: Subjective but generally aligns with public statements made historically.
– **Importance**: Unimportant – Diplomatic perspectives do not typically directly impact immediate reporting by ACLED.
– **Relevance**: False – Does not directly relate to the questioned death toll.
6. **Fact**: The U.S. provides ongoing military assistance to Ukraine, including Patriot Air Defense Systems and armored vehicles.
– **Correctness**: True, reflects ongoing military support arrangements.
– **Importance**: Unimportant – Military aid context does not impact direct death reporting for September 2025.
– **Relevance**: False – Does not alter death counts directly.
7. **Fact**: Continued civilian casualties, including deaths of children, are highlighted in UN assessments.
– **Correctness**: True, corroborated by ongoing conflict reports.
– **Importance**: Important – Civilian casualties contribute directly to the overall death toll estimate.
– **Relevance**: True – Directly relevant to the conflict casualty understanding.
8. **Fact**: Joint military exercises “Zapad 2025” by Russia and Belarus have raised alarms in Eastern Europe.
– **Correctness**: True, consistent with typical geopolitical developments.
– **Importance**: Unimportant – While potentially impactful long-term, they do not change September 2025 death reports.
– **Relevance**: False – This fact doesn’t immediately influence the reported death toll.
By evaluating these facts, focus should remain on items directly affecting combat and civilian death tolls for assessing overall state-based conflict deaths as per the original query.
Fact-checked summary:
In assessing the state-based conflict deaths in Ukraine reported by ACLED for September 2025, the pertinent and accurate facts include the ongoing occurrence of civilian casualties, which are corroborated by UN assessments and directly contribute to the overall death toll estimate, reflecting the scale and impact of the conflict. Another relevant piece of information is the recent high-intensity military engagements, such as the large-scale attack by Russia involving drones and missiles across multiple regions, albeit with varying specific figures in reports, which illustrate the overall context and operations in the conflict. While not detailing exact death counts, these facts provide important context for understanding the ongoing dynamics and casualty calculations in the conflict.
OpenAI gives a base rate of number of reported deaths (Carolyn‘s note: OpenAI failed to return a value.)
The question difficulty is rated 8 (0 to 10)
Historical weighted factors include:
Average monthly deaths in 2024, 0.3
Recent escalation trends in 2025, 0.4
Reporting capability of organizations like ACLED, 0.3
A Bayesian calculation could be performed as follows:
P(Death Toll in Range | Data) = [P(Data | Death Toll in Range) * P(Death Toll in Range)] / P(Data). Using historical factors, we assume increased likelihood for higher ranges given current trends in conflict escalation and reported attacks.
Bayesian base rate: 0.6 (Carolyn‘s note: In cases where the Multi-AI Oracle is only forecasting a single percent probability, this Bayesian rate is easy to understand. But in this case, with five probability buckets, this apparently is an overall fudge factor across all buckets.)
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: 30
The following were considered in order to produce this base rate:
The base rate balances historical monthly death tolls with recent escalation trends in 2025. Key factors include the frequency of high-intensity military engagements and the robustness of monitoring by ACLED and UN. Given ongoing high-intensity conflicts, a higher death toll range appears more likely.
Ideally, the news feed would contain the following sorts of information for a better forecast:
More precise casualty numbers, additional context on military engagements throughout September, and updates on potential peace talks or other mitigating events would aid forecasting significantly.
Some potential divergent considerations that might affect the base rate:
Potential for decreased hostilities or successful peace negotiations, improved civilian protection measures, or inaccuracies in reporting could lead to lower casualty figures than the base rate estimate.
The following chain of events are necessary for the question to resolve positively:
Continued high-intensity military engagements in Ukraine throughout September 2025 Very likely – Effective reporting and tracking of civilian and military casualties by ACLED and UN Likely – Escalation of conflicts in key regions leading to increased casualties Possible
Querying Claude (AI predicts: [“Less than 500”: 0.05, “Between 500 and 1000”: 0.2, “Between 1000 and 1500”: 0.35, “Between 1500 and 2000”: 0.25, “Greater than 2000”: 0.15] – confidence: 4)
Querying Mistral (AI predicts: [“Less than 500”: 0.05, “Between 500 and 1000”: 0.15, “Between 1000 and 1500”: 0.3, “Between 1500 and 2000”: 0.35, “Greater than 2000”: 0.15] – confidence: 5)
Querying OpenAI (AI predicts: [“Less than 500”: 0.1, “Between 500 and 1000”: 0.2, “Between 1000 and 1500”: 0.25, “Between 1500 and 2000”: 0.25, “Greater than 2000”: 0.2] – confidence: 6)
Question Type: Multiple Choice
Confidence: 5
MAPD: Avg: 0.046666666666667, Max: 0.066666666666667
# LLM responses: 3
Explanations of the above statistical measures here —>
Model value:
Less than 500: 5%
Between 500 and 1000: 21%
Between 1000 and 1500: 32%
Between 1500 and 2000: 26%
Greater than 2000: 16%
The predictions regarding the death toll in Ukraine’s ongoing high-intensity conflict are based on historical trends, the recent escalation in military engagements, and robust reporting by ACLED and UN monitoring systems. The established patterns suggest a high likelihood of casualties in the 1000-2000 range, given sustained hostilities and large-scale attacks. The reliability of the casualty tracking is supported by comprehensive methodologies, but uncertainties remain due to potential changes in reporting accuracy, possible sudden de-escalations, improved civilian protections, or shifts in military strategies. These factors might lead to deviations from the predicted ranges, highlighting the inherent unpredictability in conflict zone data collection and forecasting.
Runtime: 162 seconds.
Past forecasts by Phil’s and Jeremy ’s bots —>
Jeremy’s Ukraine forecasts are a side experiment with the VIEWS competition. The VIEWS Ukraine dashboard is pictured below. (Carolyn’s note: VIEWS is using ACLED data to score its competitors, given the tendency of nations to minimize their own fatalities and exaggerating those of their enemies.)
Source: VIEWS machine forecasting competition https://data.viewsforecasting.org/
Below, from the Institute for the Study of War and American Enterprise Institute, dated Sept. 11, 2025: https://storymaps.arcgis.com/stories/36a7f6a6f5a9448496de641cf64bd375
