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: 6%
Between 500 and 1,000: 22%
Between 1,000 and 1,500: 33%
Between 1,500 and 2,000: 28%
Greater than 2,000: 11%
7–8 minutes read
Obtaining outside data.
Checking internet sources, using keywords: Ukraine conflict deaths ACLED September 2025
News from various sources:
I’m sorry, but the content provided doesn’t contain any relevant information about the topic “Ukraine conflict deaths ACLED September 2025.” It mostly consists of repetitive phrases and lacks any substantive details. If you have more detailed content or specific questions, please provide additional text or inquire further, and I’ll be happy to help!
(Carolyn‘s note: Today the Multi-AI Oracle solved the problem of today’s failure of its news gathering model — Perplexity– by repeating its forecasts for the month of August 2025.)
Fact-checking news summary:
Given the summary provided, let’s extract specific facts, evaluate their correctness, and assess their importance and relevance.
1. **Fact**: The content provided does not contain any relevant information about the topic “Ukraine conflict deaths ACLED September 2025.”
– **True/False**: True.
– **Importance**: Important. This fact is significant because it addresses whether the content contains the needed information relevant to the question.
– **Relevance**: True. This directly affects the resolution of the question by indicating that the needed data is absent.
2. **Fact**: The content mostly consists of repetitive phrases and lacks any substantive details.
– **True/False**: True.
– **Importance**: Important. This detail helps understand why the content does not help answer the question.
– **Relevance**: True. By clarifying the nature of the content, it confirms the absence of relevant information.
These are the specific facts extracted from the summary provided. Both facts primarily confirm that the initial content was insufficient, emphasizing the need for additional data or questions to address the original query.
Fact-checked summary:
The provided summary confirms the absence of any relevant information about state-based conflict deaths in Ukraine for September 2025 as reported by ACLED. It highlights two key facts: firstly, the content does not contain any relevant details about the topic, which is important because it indicates that the needed information is missing. Secondly, the content lacks substantive details and consists mostly of repetitive phrases, confirming its irrelevance to addressing the original question. Both facts underscore the insufficiency of the initial content and the necessity for additional data to properly answer the query.
OpenAI gives a base rate of probability (Carolyn’s note: OpenAI failed to return a value.)
The question difficulty is rated 7 (0 to 10)
Historical weighted factors include:
Average monthly conflict deaths in Ukraine in previous years, 0.4
Current conflict intensity and military activity, 0.4
Reporting capability of ACLED and other organizations, 0.2
A Bayesian calculation could be performed as follows:
To compute the Bayesian probability, we multiply the probabilities derived from historical factors, adjusting for current unresolved news. Historical average monthly deaths provide a prior, current conflict intensity is used to gauge if the situation is escalating or stabilizing, and reporting capacity reflects the accuracy of data collection.
Bayesian base rate: 0.55 (Carolyn’s note: In cases where the Multi-AI Oracle is only forecasting a single percent probability, this is pretty good. But in this case, with five probability buckets, this apparently is an overall confidence level across all buckets.)
Sufficient news to provide a good forecast? 0 (0 or 1)
News is relevant, topical and unbiased? 0 (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 of 0.5 was chosen by considering historical conflict death levels in Ukraine and the recent reporting standards by ACLED, as well as the typical level of military engagement at this time of the year.
Ideally, the news feed would contain the following sorts of information for a better forecast:
For an accurate forecast, data on current hostilities, troop movements, and casualty reports are needed. Government statements or independent assessments of the situation could provide more clarity.
Some potential divergent considerations that might affect the base rate:
Factors such as sudden escalations in conflict levels, changes in international intervention, or significant reporting delays could significantly alter the expected death count for September 2025.
The following chain of events are necessary for the question to resolve positively:
Collection and reporting of state-based conflict death data in Ukraine for September 2025. Likely – Effective and comprehensive data-gathering on deaths by relevant organizations such as ACLED. Likely
Querying Claude (AI predicts: [“Less than 500”: 0.15, “Between 500 and 1000”: 0.25, “Between 1000 and 1500”: 0.3, “Between 1500 and 2000”: 0.2, “Greater than 2000”: 0.1] – confidence: 4)
Querying Mistral (AI predicts: [“Less than 500”: 0.05, “Between 500 and 1000”: 0.2, “Between 1000 and 1500”: 0.35, “Between 1500 and 2000”: 0.3, “Greater than 2000”: 0.1] – confidence: 4)
Querying OpenAI (AI predicts: [“Less than 500”: 0.05, “Between 500 and 1000”: 0.15, “Between 1000 and 1500”: 0.25, “Between 1500 and 2000”: 0.25, “Greater than 2000”: 0.3] – confidence: 6)
Question Type: Multiple Choice
Confidence: 4
MAPD: Avg: 0.08, Max: 0.13333333333333
# LLM responses: 3
Explanations of the above statistical measures here —>
Model value:
* Less than 500: 6%
* Between 500 and 1,000: 22%
* Between 1,000 and 1,500: 33%
* Between 1,500 and 2,000: 28%
* Greater than 2,000: 11%
The various AI systems base their predictions of conflict deaths in Ukraine on historical data from ACLED and the ongoing nature of the Russia-Ukraine conflict, emphasizing patterns of moderate to high-intensity fighting with monthly death tolls ranging between 1,000 and 2,000. Factors such as the stability of the front lines, geopolitical forces, and seasonal patterns influence these predictions, with September historically depicting moderate activity levels. However, the AI systems also acknowledge the potential for inaccuracies due to unforeseen military escalations or de-escalations, changes in tactics, and potential delays or inaccuracies in data reporting. The unpredictable nature of war and the possibility of diplomatic breakthroughs add further layers of uncertainty, making it challenging to draw definitive conclusions solely from historical data.
Runtime: 68 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 Aljazeera https://www.aljazeera.com/news/2025/8/28/russia-advances-in-kupiansk-rejects-us-push-for-zelenskyy-putin-summit
