Ukraine, Palestine/Israel, Sudan, Pakistan and Nigeria are projected to see the highest battle-related death tolls in 2026, according to a world-leading AI-driven conflict forecasting system.
The Violence and Impacts Early-Warning System (VIEWS) is an award-winning conflict prediction model developed by the Peace Research Institute Oslo (PRIO) and Uppsala University, that produces monthly forecasts for violent conflict up to three years ahead. Using machine-learning techniques and a large-scale data pipeline, it anticipates where and how severely armed conflict is likely to escalate.
According to the latest VIEWS forecast for state-based armed conflict, the estimated number of battle-deaths in 2026 is: *
- Ukraine: 28,300
- Palestine/Israel: 7,700
- Sudan: 4,300
- Pakistan: 2,000
- Nigeria: 1,900
- Ethiopia: 1,800
- Somalia: 1,700
- Syria: 1,400
- Yemen: 1,300
- Burkina Faso: 1,200
In Palestine/Israel,** the model projects fewer deaths in 2026 than the roughly 14,000 state-based conflict deaths recorded by the Uppsala Conflict Data Program between January and October 2025, partly reflecting the Gaza ceasefire agreed in October. In Ukraine, the 2026 projection is also below the 59,600 recorded deaths so far in 2025.*** While the model incorporates the possibility that both conflicts could end in 2026, VIEWS’ forecasts tend to err on the conservative side.
The predicted level of violence in Sudan remains below the approximate 7,200 fatalities recorded by UCDP in 2025 to date. However, the model’s projected 2026 death toll for Sudan has more than doubled in the past month, underscoring the rapid deterioration of the security situation.
“Our forecast highlights where conflict intensity is most likely to surge next year, giving governments, the UN and aid organizations a stronger basis for early planning and saving lives,” said Håvard Hegre, the PRIO Research Professor who leads the VIEWS team. “It should serve as a clear warning bell for anyone working on Sudan that the situation on the ground may get much worse in the coming months.”
The VIEWS model has a strong predictive track record, correctly identifying seven of the ten deadliest countries in 2024 and six of the top ten in 2023.
The projected fatality numbers reflect the model’s central estimate – or “main mean” – for state-based conflict deaths in 2026. Its architecture integrates extensive open datasets, including historical violence, socioeconomic indicators, climate conditions, geography and political institutions.
“These forecasts should be seen as best-estimate scenarios, not certainties. Many forces – including diplomacy, peacebuilding, humanitarian action and geopolitical intervention – can shift conflict trajectories. Nevertheless, the results can support actors on the ground in prioritizing attention and resources to where they are likely to be needed most,” said Hegre.
For more information or to arrange an interview
- Contact Michelle Delaney, Communication Director at the Peace Research Institute Oslo (PRIO) | michelle@prio.org | mobile +47 941 65 579.
- To read more about the Violence and Impacts Early-Warning System (VIEWS) prediction model, go to their website.
Notes
- * Estimated and recorded battle-death figures reported above are rounded to the closest hundred people killed. Predicted numbers are from VIEWS’ latest model, “fatalities003”.
- ** The forecasts for Palestine/Israel include forecasts for Gaza and the West Bank, as well as fatalities related to the conflict in Israeli territory.
- *** The data on fatalities in 2025 is from Uppsala Conflict Data Program’s Candidate datasets, which have been pre-processed by VIEWS for use in their model. The values listed here may therefore differ slightly from the original data. Click here to learn more about the original datasets.
- Click here to access VIEWS’ adaptation of the data.
