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 June, 2025?
Here’s what Jeremy’s Multi-AI Oracle predicts today:
* Less than 500: 18%
* Between 500 and 1000: 29%
* Between 1000 and 1500: 29%
* Between 1500 and 2000: 18%
* Greater than 2000: 6%
Obtaining outside data.
Checking internet sources, using keywords: Ukraine conflict deaths June 2025
Results from various sources: (Carolyn‘s note: News retrieval failed today. As a consequence, its forecast today is much lower than on June 3, 2025.)
Fact-checking news summary:
Please provide the summary that I should analyze. Once you share the summary, I’ll be able to list the specific facts and evaluate them according to your criteria.
Fact-checked summary:
I’m sorry, but I can’t provide specific numbers for future events, such as conflict deaths reported by ACLED in June 2025, since it involves predictions about future data. Additionally, the information regarding future ACLED reports is not available in current datasets, as my training only includes data up to October 2023.
Couldn’t get base forecast
Querying Claude (AI predicts: [“Less than 500”: 0.15, “Between 500 and 1000”: 0.35, “Between 1000 and 1500”: 0.3, “Between 1500 and 2000”: 0.15, “Greater than 2000”: 0.05] – confidence: 4)
Querying Mistral (AI predicts: [“Less than 500”: 0.6, “Between 500 and 1000”: 0.25, “Between 1000 and 1500”: 0.1, “Between 1500 and 2000”: 0.04, “Greater than 2000”: 0.01] – confidence: 6)
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”: 0.3, “Greater than 2000”: 0.25] – confidence: 6)
Question Type: Multiple Choice
Confidence: 6
MAPD: Avg: 0.19333333333333, Max: 0.36666666666667
# LLM responses: 3
Explanations of the above statistical measures here —>
Model value:
* Less than 500: 18%
* Between 500 and 1000: 29%
* Between 1000 and 1500: 29%
* Between 1500 and 2000: 18%
* Greater than 2000: 6%
The predictions about conflict fatalities in Ukraine for June 2025 are primarily based on historical data, suggesting a range of 500-1500 deaths as most probable, with lower chances for extreme values. The reasoning includes past patterns of decreased intensity and seasonal trends that typically feature moderate activity. However, the analyses acknowledge the inherent unpredictability of conflict zones. Factors like ongoing diplomatic efforts hint at potential de-escalation, yet recent trends and continued hostilities suggest casualties could remain significant. There are several factors that could disrupt these predictions, including sudden military escalations, changes in political leadership, external influences, and reporting inaccuracies due to ACLED’s methodology or challenges in data collection. Hence, while the prediction relies on established conflict dynamics, it remains susceptible to unexpected geopolitical or tactical shifts.
Runtime: 226 seconds. (Carolyn’s note: Despite the failure of Jeremy’s Multi-AI Oracle to access current news about Ukraine, this run required twice the run time that is normal. One possible cause for this weird failure mode is that some of the big generative AI companies have been cutting back on performance of their low cost and free models. They have been heavily subsidizing their models. Now they need to experiment with getting customers to pay more. Another problem is that these models often crash . You can see examples of these crashes in Phil Godzin’s full responses, including for today’s results for Ukraine here —>)
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/
