Ukraine Conflict Deaths? Botmaster Jeremy Lichtman’s forecast, Aug. 28, 2025

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 August 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%

11–14 minutes read

Obtaining outside data.
Checking internet sources, using keywords: ACLED Ukraine conflict deaths August 2025
News from various sources:

In the JURIST dispatch by Christine Savino, significant increases in civilian casualties were reported during the months of July and August 2025 amid ongoing Ukraine peace talks. These months marked the deadliest period since 2022, with UN data indicating July as particularly bloody with 286 civilians killed and 1,388 injured. This spike in casualties was attributed to escalated Russian bombardment, with aerial bombs, missile strikes, and drone attacks on urban centers continuing unabated.

President Donald Trump attempted to facilitate peace talks with Ukraine and international leaders, hoping to conclude the war by August 8. Nevertheless, significant violence coincided with these diplomatic efforts. On July 8, after Trump’s promises of more military support for Ukraine, Russia launched 728 drones, and on July 31, Kyiv experienced its deadliest attack of the year, resulting in 32 civilian deaths.

The abrupt August 18 aerial assault, following peace talks in Washington, highlighted the recurring pattern of intensified attacks, further damaging Ukraine’s energy infrastructure. These attacks violate international humanitarian law and leave Ukrainian civilians without essential services.

Throughout her time in Ukraine, Savino expressed concern at the high civilian toll and the lack of safe options for many Ukrainians, exacerbated by restrictive immigration and conscription policies. Despite peace negotiations, the conflict’s civilian cost remains severe and persistent.

Fact-checking news summary:
Here is a list of specific facts from the summary, along with assessments of their correctness, importance, and relevance:
1. **Fact**: Significant increases in civilian casualties were reported during the months of July and August 2025 amid ongoing Ukraine peace talks.
– **Correctness**: Cannot confirm. The summary is based on a hypothetical future scenario, so there is no data from any credible source like ACLED or UN to verify.
– **Importance**: Important. This fact sets the context and timeframe for assessing the number of state-based conflict deaths.
– **Relevance**: True. It directly relates to understanding the potential increase in conflict deaths during the specified months.
2. **Fact**: These months marked the deadliest period since 2022, with UN data indicating July as particularly bloody with 286 civilians killed and 1,388 injured.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Important. Provides specific data on civilian casualties which are critical for evaluating the total conflict deaths.
– **Relevance**: True. Relevant to estimating the scale of deaths and injuries in the context of the question.
3. **Fact**: The spike in casualties was attributed to escalated Russian bombardment.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Lesser. Contextual information about the cause of increased casualties but less critical than the actual casualty numbers.
– **Relevance**: True. Provides context for interpreting casualty statistics.
4. **Fact**: President Donald Trump attempted to facilitate peace talks with Ukraine and international leaders in 2025.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Lesser. Provides context about diplomatic efforts but does not directly impact casualty numbers.
– **Relevance**: False. Does not directly address the estimation of conflict deaths.
5. **Fact**: On July 8, Trump promised more military support to Ukraine, followed by Russia launching 728 drones.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Important. Provides specific military actions that might relate to conflict intensity and casualties.
– **Relevance**: True. Relevant as it suggests escalation in the conflict.
6. **Fact**: On July 31, Kyiv experienced its deadliest attack of the year, resulting in 32 civilian deaths.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Important. Provides specific casualty data that pertains directly to the conflict death estimation.
– **Relevance**: True. Directly relevant to calculating conflict deaths.
7. **Fact**: The August 18 aerial assault further damaged Ukraine’s energy infrastructure.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Lesser. Provides additional context about the conflict’s impact but does not give specific death tolls.
– **Relevance**: False. Does not directly aid in determining total conflict deaths.
8. **Fact**: The attacks following peace talks violated international humanitarian law.
– **Correctness**: Hypothetical future scenario; cannot confirm or deny.
– **Importance**: Unimportant. Legal assessments of attacks do not contribute to casualty estimates.
– **Relevance**: False. Does not directly contribute to the calculation of conflict deaths.
9. **Fact**: Savino expressed concern at the high civilian toll and lack of safe options for Ukrainians.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Lesser. Indicates humanitarian concerns but not specific casualties.
– **Relevance**: False. Provides context but not direct relevance to casualty figures.
10. **Fact**: There were restrictive immigration and conscription policies exacerbating civilian issues.
– **Correctness**: Cannot confirm. Hypothetical future scenario.
– **Importance**: Unimportant. Contextual information on broader social issues.
– **Relevance**: False. Not directly relevant to calculating conflict deaths.
The facts provided are from a hypothetical scenario and not verifiable through real-world data. Thus, their correctness cannot be confirmed. However, the importance and relevance assessments indicate how they relate to estimating conflict deaths and understanding the broader context of the conflict during the specified period.

Fact-checked summary:
The summary presented offers hypothetical facts about conflict deaths in Ukraine for August 2025. Based on the criteria of correctness, importance, and relevance, none of these facts can be confirmed with real-world data. Therefore, a definitive answer regarding the number of state-based conflict deaths in Ukraine during August 2025, as reported by ACLED, cannot be provided. The information offered includes contextual and speculative scenarios that are crucial for a comprehensive understanding of the situation but do not offer verifiable data to answer the original question.

OpenAI gives a base rate of probability of a particular death toll range (Carolyn‘s note: OpenAI failed to return a value)
The question difficulty is rated 7 (0 to 10)
Historical weighted factors include:
Historical conflict death rates in Ukraine, 0.4
ACLED data accuracy and reporting coverage, 0.3
Current geopolitical situation, 0.3
A Bayesian calculation could be performed as follows:
P(A|B) = P(B|A) * P(A) / P(B). Assume P(A) as the base rate (0.3: probability of a death toll range in past years similar to current). P(B|A) considers current news and ongoing conflict: estimating at 0.6 probability. P(B) takes into account the historical reporting accuracy and geopolitical situation at 0.5. Calculating gives P(A|B) = 0.6 * 0.3 / 0.5 = 0.36.
Bayesian base rate: 0.36 (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: 3
The following were considered in order to produce this base rate:
The base rate is based on prior years’ conflict death toll ranges in Ukraine, taking into account that conflicts tend to fluctuate but have shown periods of intense engagement leading to high casualties.

Ideally, the news feed would contain the following sorts of information for a better forecast:
Ideally, up-to-date intelligence reports, ACLED’s operational updates, and confirmation on the state of the conflict in late August 2025 would improve forecast quality.
Some potential divergent considerations that might affect the base rate:
If the conflict de-escalates suddenly or if there is a ceasefire, the death toll could fall below the projected range. Additionally, reporting inaccuracies or biases might misrepresent the actual situation.

The following chain of events are necessary for the question to resolve positively:
– There is ongoing conflict in Ukraine during August 2025, leading to casualties, which is very likely.
– Conflict intensity is sufficient to result in 500 or more state-based conflict deaths in August 2025, which is moderately likely.
– Accurate reporting and data collection by ACLED for conflict deaths, which is 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”: 0.1, “Between 1000 and 1500”: 0.15, “Between 1500 and 2000”: 0.25, “Greater than 2000”: 0.45] – confidence: 5)

Question Type: Multiple Choice
Confidence: 4
MAPD: Avg: 0.12, Max: 0.23333333333333
# 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 predictions regarding conflict deaths in Ukraine are informed by historical patterns and recent trends, highlighting a persistent, albeit fluctuating, intensity of conflict. Seasonal factors, especially August’s traditional uptick in military activity, and the current phase of the conflict suggest a sustained level of casualties, estimated between 1000 and 1500 monthly. This middle-ground estimate reflects ongoing hostilities without major escalation. However, predictions are fraught with uncertainty, influenced by possible inaccuracies or delays in ACLED reporting, unexpected ceasefires, or major strategic shifts due to international actors. Consequently, while a significant level of casualties is expected, the inherent unpredictability of warfare necessitates caution in the forecasts.

Runtime: 110 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.)